I followed the hype: Reddit, Slashdot’s front page, months of thumbs up on my blog and various video forums by Linux users for OpenShot. Given that I’m longing for a usable Linux video editor since 2003, and given that OpenShot version 1.0 had just been released, I naturally gave it a go, by also downloading its provided dependencies on my Ubuntu Linux 9.10.
The provided .deb files installed without a hitch. When I loaded the app, I was provided with a pretty traditional and easy to understand video editing playground: files, transitions, effects on one side, a preview window on the right, and the timeline on the bottom with tools such as “cutting”. Easy to gasp, and very easy to use.
Right-clicking on the clips in the timeline will load a dialog where you can control speed, plugins, and more. Transitions are working properly if you put the clips on different tracks, but they didn’t do what I needed when I overlapped two clips and placed a transition on top. Additionally, the “Video Preview” tab, the preview and timeline toolbar icons, and the “Timeline – Sequence 1” tabs take way too much space. On an XGA monitor, the preview screen gets reduced to a stamp-size video, while PiTiVi fairs way better in that department. The developer of OpenShot must optimize his interface to not take unnecessary space.
Another gripe is the fact that you can’t set the aspect ratio of your footage and de-interlacing project-wide. There are a number of templates, but the project properties and exporting dialog don’t offer enough flexibility. There’s also a bug in the AppleTV exporting dialog where it exports at 720/30p, while the AppleTV can’t do more than 720/24p.
Except “blur”, “brightness”, “deinterlace”, “gamma”, and “white balance”, most of the rest of the effects are not very useful. Just like any other Linux video editor out there it’s using the frei0r third party plugin system, which are basic, and buggy. Additionally, speaking for me, the real value of post-processing footage is in color grading, not in these silly and super-cliche spiral transitions that are so ’80s. Where are the 3-way color correctors, contrast, color isolation, etc?
Additionally, some more advanced features, like keyframes, are architected in a non-useful way. Instead of being able to use keyframes within a clip, we can only use them at beginning or end, and they won’t work with plugins, but only with position of the clip in the preview plane. Similarly, the “slicing” options, which automatically cuts into pieces a clip in the timeline, is missing the point: it doesn’t scale, usability-wise.
I tried every possible format out there: XViD, MKV, h.264 in MP4/MOV/MTS, .m2ts, m2t, mjpeg AVI/MOV, OGV etc. The application and its custom ffmpeg it used was able to load all file formats I threw at it, but it was super slow. Previewing speed while on the media bin was as expected, but when dragged in the timeline I couldn’t find a single instance where the video would preview smoothly. Even a 320×176 OGV file was bleeding framerate. In addition, I was able to reproducibly crash the application using a WMV VGA video, and a 720/30p MJPEG AVI from the latest Olympus dSLR.
In addition to terrible video performance, there was the audio that was misbehaving: there was audio crackling during ALL videos. And when I imported just a single mp3 file, there was still major crackling.
I’m afraid that doing actual work with that editor is out of the question: I’d get about one lockup every minute! Application freeze after freeze. Previewing videos either in the timeline or not it would lock up the app quite easily, and in some case it locked up my whole X11 (had to kill X). At one instance I even locked up the app by trying to open my Desktop folder to load a .mov file (a file that earlier did work with the app). Previewing, adding clips, adding plugins or transitions, all eventually resulted in either freezing or crashing (and no, there’s nothing wrong with my PC or Ubuntu installation). Freezing in particular was so common, that I’m unable to write a more traditional review where the process follows a specific editing workflow — simply because editing anything serious was not really feasible. And when I tried to just export a small clip, it exported out the file (Vimeo HD template), but the app crashed right on the last step so the produced file was not really playable.
Overall, I have to say that this is the most problematic application I have encountered in the last 3-4 years. This is not an 1.0 application. It’s not even beta. It’s alpha quality — at best. It is infinitely more unstable than either KDEnLive or PiTiVi (editors that don’t fair well either, but they’re like bright stars compared to OpenShot).
I’m not trying to trash 1.5 years of work for the main developer. I’m sure it was hard work, and I know he’s a cool guy in the few times we have corresponded during the past year. Plus, I know very well how hard it is to write a video editor that actually works as expected (hint: more complex than developing something like Firefox). But calling this 1.0 is such an unimaginable stretch that makes me want to pull my hair out. Sure, you can argue all you want that “1.0” is just an arbitrary number, but when this release is so over-hyped and drives consumers to use this application, then I have a problem with the situation. As for those who will say “well, try it on another Linux distro”, my answer is “no”. Ubuntu 9.10 is very popular, and these packages should have been tested thoroughly before they were released for general consumption.
I hope the developer can find the mental strength to download A LOT of footage from many different cameras and formats, and start testing, testing, testing. Forget about new features for a YEAR. Just test and optimize the darn thing like there’s no tomorrow. And then call it 1.0.1.
Rating: 3/10
I had a very similar experience. I have a Panasonic DMC-ZS3 that shots 720p movies in standard AVCHD format. I did quite a few clips recently, postponed editing them because kdenlive was crashing on me all the time… So I was waiting eagerly for something better…
Downloaded OpenShot last week, tried it immediately. First of all, I can’t just drag my video files to the Project Files window (but that’s probably GTK’s fault, I use KDE). At least you can select multiple files in the file selector. Then I was looking for something really simple – fade from black & and fade to black in between clips. I wanted to stich together about 15 short clips this way… Well, no luck – instead we have fractal 1, 2, 3, 4 and lots and lots of useless transitions.. Than it was slooow – just like Eugenia described it.
Well, basically this is what I wanted (that’s a kdenlive screenshot)
http://picasaweb.google.com/lh/photo/_uCiI4KkI2-Pm4aaF3OrxA?feat=di…
Now that’s what I call a really basic functionality, and yet, I couldn’t do it!
It also took me some time to find fade in and fade for SOUND – it was hidden in the video properties dialogue. Why on earth did they put it there? Probably because that seems to be the only sound effect they support… In kdenlive you have a drop-down menu where you can select all effects, video effects and audio effects (plus custom).
Anyway, for the past three days I’ve been using kdenlive again – and it’s been rock stable – not a single crash, finished two short videos (composed of about 15-20 clips) already. I don’t know exactly when it became stable – last time I tried to do what I’m doing now was about 2 months ago. It might be an update since then, or may be disabling compositing helped, but the thing is, OpenShot has nothing to offer compared to kdenlive. It has less functionality, which would be fine if it was simpler to use, but it’s not. I don’t need all the functionality kdenlive has (although they might come in handy later), and OpenShot really looked promising, but it does not live up to the hype at all.
What’s your opinion about this new project of VLC:
http://vlmc.org/
Looks interesting to me.
From what I’ve seen from it so far, is that it’s progressing in the wrong way: two preview panels are old-school. Also, I don’t think I can wait another 5 years for it become stable (since that project is still in its infancy). So I don’t consider it a contender.
As far as I’m concerned, if you want to do some real work with video editing, even just as an amateur-but-serious videographer, you still need to use Windows or Mac. It’s unfortunate, but I see *useful* video editing on GNU/Linux as the final frontier for the platform. I’m still “on the waiting”.
5 years may be true for this project, but not because it is in its infancy. The last version of iMovie took a few months to develop by one software engineer, this was all mentioned when Steve Jobs demonstrated the app at its premiere. There may or may not be a lesson in all of that for the Linux community.
Please, don’t believe the hype that Jobs is selling. The newer type of iMovie was not developed by a single engineer. It might have _started_ by a single engineer as a proof of concept, but to get the release out, the whole team was needed. And besides, that engineer(s) had a solid framework to work on, Quicktime, and it is already known that the new iMovie reused code from FCP. So it was not all from scratch. Add to that the fact that the first version of the new iMovie broke ALL compatibility with third party plugins (in fact, it didn’t have a plugin system), and even iDVD support, among many other things, that pissed off a lot of people. There were a lot of angry people out there when the new iMovie was released, and that’s why Apple continued giving away the 2006 version too.
I know about Jobs and the reality distortion field. But I do not recall anything in his statements that said it was a proof of concept that the single developer made, but was in fact the application that shipped. It was the application that was developed by a single developer not a prototype, while the team continued to develop the iMovie HD version, which was dropped at the last minute in favor of the newer version. I have no reason to believe that Jobs was lying other than your claims. Is there is an article or some comment from someone at Apple that your citing?
I was one of those very very pissed off people.
To this day, the new iMovie (yes, even the 09 version) is inferior in many ways to iMovie HD.
It does have a few things going for it (more like iPhoto in-app database) but I stuck with HD as long as I could… which ended when I was given a USB video camera, which iMovie HD could not operate with.
I miss some of the multi-track features. But I miss speed adjustment, playback direction (forward / reverse), and in-app multi track audio editing the most.
At least, I haven’t figured out how to do this all with the new iMovie…
And the playhead controls in the new iMovie are DAMN NEAR UNUSABLE. PAIN IN THE ASS!
First, let me say thank you for your review.
It is really painful to honestly describe the state of multimedia procuction applications on Linux (and thus pretty much all Unices) today. I mean, after all those years…
Not that it should suprise anyone: This is difficult, demanding and time consuming work. OSS can’t really compete even with most entry level video editing programs because the experts are working full time at a company and get paid. And if the boss says the dirty work has to be done (like hunting all those bugs), then employee has to.
The ugly thing is the people who always say that there is no problem on Linux with availability of good multimedia production apps. As soon as you say ‘music’, somebody will come along and say Ardour or Rosegarden. In 3D it’s Blender, in image editing it’s GIMP, in DTP it’s Scribus, in DVD authoring it’s $POSTER_CHILD_APP_X, etc…
But when you look a bit closer or actually try to do something with these apps, you often very soon encounter problems. May it be UI, lack of features, lack of stability, poor performance or incompatibilities. This is my experience and the review plays into my view. Although not everything is as bad as OpenShot.
Edited 2010-01-11 09:13 UTC
Well, five years ago I would have agreed with you, although multimedia on Linux has come leaps and bounds since then. I’m a professional entertainer, and I have been using nothing but oss for the last three years. I have made 6 albums on Ardour (though I’ve never used much midi), several videos on Cinelerra, and countless posters, packaging, and merchandise designs with GIMP and Inkscape. Now, there certainly were things that I had to relearn when switching from Windows/Warez (when I started out my business showed a net loss, so there’s no way I could have afforded Cubase, Adobe CS, or ProTools), but in the end a person can do anything on Linux that can be done (albeit in perhaps a more user-friendly way) on Windows. I’ve actually found Mac to be my particular sore spot. While I love some of the software that you can get only on a Mac (Logic and FCP to name a couple), in general you end up either spending twice as much as MS folks for the same functionality (and warez is harder to find), or there are no free options if you decide you want to go that route (unless you want to compile everything from ports).
But I digress, the whole point of the article is that Openshot is not ready for 1.0, which I totally agree with. And although I love Cinelerra feature-wise, I have to concede that it’s probably the ugliest UI I’ve ever seen in a piece of software that wasn’t made pre-1994 (not to mention that it’s a pain in the ass to compile).
Don’t believe that you can do multimedia with oss? check out dickmacinnis.com. All FOSS (even the website – made with Kompozer).
No offense, but your website looks like ass. And given the fact that the content is largely courtesy of Google and ReverbNation, I really wouldn’t put it forth as an example that you also can make websites with FOSS.
Besides, you miss the point. I don’t think that anyone doubts that you can produce multimedia with FOSS. The trouble is that you just can’t do it well. FOSS for multimedia production may be good enough, but good enough almost always doesn’t cut it in the real world, which is the place that doesn’t exist according to fervent FOSS advocates.
If you’re a semi-pro or a pro that earns his living with multimedia production of any kind, then why would you be dumb enough to endure the pain of dealing with FOSS multimedia production software? If someone, for example, wants to break into Visual Effects and get a compositing job, is she ill-advised to learn by using Blender or Cinelerra? Hell yes! Instead, she should use the tools that are used in the business, like Combustion or Nuke, both of which are available as Learner Editions at no cost.
I really hate this ubiquitous tagline: “You can do that with FOSS, too!” At least the greater half of all FOSS software seems to be made with this in mind. The end-result: glorified, but horrible proof-of-concepts. So, sure, you can somehow produce multimedia content with FOSS, but who would want to?
OK I checked that link. Why am I thinking Atari Teenage Riot could do that on an Amiga 500 .. err 20 years ago? Because they could ! ! ! — And did.
Way to go dude… btw check ATR’s vids out.. (before you flame – lol)
uh.. it isn’t a 500 – it is an ST, of course, my bad…
The big studios predominantly use Linux and OSS programs in post production imaging.
For example, the open source nature of Cinepaint is the main reason why the major film studios and major animation houses use it, and not Photoshop. Not only is the development a lot faster (having 32-bit color depth years before Photoshop), but they can (and do) accelerate the app’s development if they need a feature.
Post production techniques are constantly developing at a rapid pace, and every minute of the average theatrical feature film costs about one million dollars to make. With that kind of money involved in every detail and with competitors writing their own image manipulation code, do you think that the studios are going to settle for the stock features of off-the-shelf proprietary programs like Photoshop? They need an image editor that they can develop in-house to their advantage. This advantage is why Cinepaint is so much more attractive to the studios than Photoshop.
Here’s the NLE/compostitor, Piranha: http://www.ifx.com/products/piranha
Piranha has been around forever on Linux. Scroll down the Piranha web page and note some of the features. One could say that Piranha is slightly more sophisticated than Final Cut Pro!
Same goes for Piranha’s little brother, Ant: http://www.ifx.com/ant
Another big advantage of Linux NLEs over FCP: one can change to a darker theme so that the OS’s window elements aren’t glaring into one’s eyeballs in a dark edit bay.
What’s wrong with Ardour? … Blender? Both are incredibly powerful programs.
GIMP is better than Photoshop in many ways. First of all, it’s easier to use. I have been able to accomplish a lot of things in GIMP, that my professional photographer friends couldn’t do in Photoshop, with years of Photoshop experience. Recently, a friend of mine (who is a pro with probably eight years Photoshop experience) could not figure out how to separately extract the images in an animated gif, so he emailed it to me. I had never done it before, but, using GIMP, I was emailing him the separate images within ten minutes of when I received the gif.
The open source nature of the GIMP is an advantage. One can fork it off into new editor if one desires — such as Cinepaint!
I can use the latest version of GIMP/Cinepaint compiled for 64-bit. Try that on a Mac with Photoshop! 😉
Furthermore, I can put GIMP/Cinepaint on a small Linux distro on a live CD (or live USB stick) and it will boot on almost any X86 PC, so I can travel with my photo editor and I don’t have to worry whether it will be comptible with the OS on the machine that I encounter.
If I use Tiny Core Linux, both the OS and GIMP combined on the live CD/USB come to only 17Mb! Try that with your proprietary software!
I can do the same with Ardour, Blender and probably with Ant and Piranha.
By the way, Photoshop relies on the open source dcraw for its Adobe Camera Raw plugin to import raw camera files (an essential function for pro photographers). Read the second sentence: http://wiki.panotools.org/Dcraw
I hear a lot of vague criticism of OSS and Linux, but when it gets down to the details, I find that it is often superior to the alternatives.
Also, give the OpenShot guy a break. He’s probably been coding his NLE for less than a year, and he’s accomplished a lot it that short time. No one should expect it to be perfect right now.
Edited 2010-01-12 01:43 UTC
Yes, many studios do. ILM runs almost exclusively on Linux. So what?
No! Just, no. If something is open source or not is not important. It’s important if it gets the job done. CinePaint obviously does, because it was developed with the VFX industry in mind by someone associated with that industry. But the job was done by someone first. Then the studios decided to use it, and, in some cases, contribute to it. But it didn’t originate at a studio.
Your whole post is full of “moving the goalposts.” Yes, the majority of the VFX industry has capitalized on Linux. Nearly every VFX software is available for Linux. But that is totally unrelated to the sorry state of consumer-grade multimedia software. Do I as a user care that ILM runs Linux? No, but I do care if the video editors available to me suck.
Edited 2010-01-12 08:02 UTC
So, the OP of this sub-thread suggests that Linux multimedia programs cannot compete with the entry level apps in the non-Linux/Unix world. If the “big boys” are using Linux instead of Windows/Mac, obviously, the OP is incorrect.
Did I really have to explain that point?
I don’t think that we disagree that much in our points, but we seem to have very different attitudes.
Cinepaint’s origins are irrelevant. The studios wouldn’t have started using Cinepaint if it couldn’t get the job done. It could, and it is open source, which is why they don’t switch to something else proprietary that could now possibly get the job done (like 32-bit depth Photoshop).
You said yourself that the studios have contributed to Cinepaint, so we agree that they have a motivation to mess with the code. However, it is most likely that a lot of their Cinepaint development stays in-house, so that that they keep an edge over their competitors. They can’t get that edge with Photoshop.
I’m not sure what the hell your quoted phrase means, but I am not very pleased that you are making personal comments.
First of all, anyone can obtain for free all of the open source software that the “big boys” are using. So there is no barrier there.
Secondly, I wholeheartedly disagree that the Linux “consumer-grade” software is in a “sorry” state. Exactly what are the problems that you have encountered? Please give specific examples.
I hope that you are not complaining about something as trivial as an unfamiliar GUI. Keep in mind that the folks using a high-end Linux NLE running 20 hyper-HD threads don’t care that the look of its window widgets is not visually consistent with the window buttons on Itunes!
You want a Linux NLE that will knock the pants off of FCP and standard-issue Avid? — get Ant!
Edited 2010-01-12 09:45 UTC
It’s not important for YOU. Being Open Source is a FEATURE. It means you can MODIFY the software to SUIT YOUR NEEDS. And don’t tell me no consumer does it, THEY DO.
Don’t quote me out of context. I was specifically referring to VFX studios. For them it is by and large irrelevant if an application is Open Source or not. They use what gets the job done most efficiently. In-house, open source, proprietary — it really doesn’t matter to them. What matters are costs and benefits.
And for the sake of god: Open Source is only then a feature, if you can leverage it, and always under the assumption that the code quality is at least above average! In your imagination every user is a hacker. Ever wondered why Linux can’t grab more of the Desktop market? No? Thought so.
Oh, yeah! This is a true classic in Linux advocacy.
Maybe you should bolster up your argument by saying also that, you know, Linux runs on supercomputers?
Maybe you should do some research, and, then, actually make a decent point.
And it is used in supercomputers, render farms, I know! You are talking about the super high end here that has soooo little volume.
If Cinepaint is so great, why isn’t it bundled in most distributions?
Again, you are talking about some people in the high end here.
Man, this is so bad. Did you try to get it? For Germany they only mentioned a swiss distributor that has prices well hidden somewhere. I couldn’t find it. This is a lot like when trying to buy other super high end and super expensive software like Maya.
Yeah, whatever…
Ardour is NOT a fully featured DAW like Pro Tools, Logic, Cubase etc. It has nothing to offer for MIDI.
Blender’s UI is very very difficult to learn. I tried it myself, now I use Cinema 4D. I bought it and never regretted! Blender was originally proprietary commercial software. If it were better, they wouldn’t be out of business by now.
It’s hard to find people like you these days. GIMP’s GUI is awkward, that’s common sense.
Why would anybody do that? You travel with your laptop which contains all the software you need. Period.
The question is just: Is time worth anything to you? I mean if mentioned OSS would be all that great, why would anybody still buy software? The answer is: You glorify OSS like there is no tomorrow. But as I said, when you look a little deeper, you find problems in the OSS ‘alternatives’.
Blender’s UI is the most productive interface on the market. It’s not difficult to learn but it takes time, yes. When you master it, you can be more productive than on any other 3D modeling software. Put one hand on the keyboard and one hand on the mouse and learn the shortcuts.
Edited 2010-01-12 10:12 UTC
Sorry, but have you ever used something different? I, at least, used Maya, Lightwave, C4D and tried Blender. If you use this type of software you’ll always have to use shortcuts a lot. And I do. So no argument here.
And if Blender is so awesome to use, why are the developers changing everything for 2.6 ??? They are a lot more honest than you are.
Yes.
Oh boy, don’t be too specific!
BTW, what about an answer to my last question?
Read their roadmap. They want all the blender interface functions to be scriptable in Python. It’s all documented on their web site.
The OP of this sub-thread was trying to suggest that Linux media software is inferior. Of course, as you admit, Linux software is quite the opposite — high end. The OP didn’t make any assertions regarding the volume of users.
It is bundled with a many of them, and it is certainly in all of the major repositories.
Exactly. Linux software is much better!
No… This is very, very good. Yes, Piranha is super high end — $200,000+. Ant is probably less expensive, maybe $10,000-$15,000.
I guess some of us haven’t been in very many edit bays.
Actually, I appears that it already has some MIDI functionality and in that regard is developing fast. Is that all?
I have used Blender a little and had no problem with the UI.
A lot of things can force a company out of business, even if the company has a great product. It happens all the time.
On the other hand, it’s very easy to find people like you, but perhaps we should refrain from personal remarks.
That is not common sense — that is a common myth.
Specific examples, please.
Ummm, maybe they don’t have a laptop. Or maybe, they don’t want to carry a laptop. Or perhaps they are traveling to a location where there is a computer which is more powerful than one’s laptop, so it would be easier/faster to use the live CD/USB, etc.
There many reasons why people currently use live CD/USBs. Period.
Again, with the personal remarks!
Time is worth much to me, and I save a lot of it by using Linux/OSS and by not having to constantly endure ignoramuses who pose questions and then follow-up with feeble and unrelated insults.
If your time is so important, why are you wasting it on inferior software, such as FCP? — spend the money and get Ant!
You have only named one concrete OSS problem — Ardour’s temporary lack of MIDI functionality. You are going to have to come up with more examples than that one, if you expect to characterize Linux and OSS as “problematic.”
By the way, there is no 64-bit Photoshop functionality on the Mac platform, so that cancels out your point about Ardour and MIDI.
Edited 2010-01-12 11:26 UTC
So there is a NLE for Linux for the cost of a new car. I didn’t know that, you score. I should have made clearer that I don’t talk about the ultra high end but something that an average joe or prosumer/hobbyist can buy.
Really? Correct me if I missed something, but what I found out is that it is very difficult do get a package for a current distribution. Even specialized ones like UbuntuStudio lack Cinepaint. This is a quote from their website, that explains why:
CinePaint was removed from Debian lenny (testing) because Debian has dropped support for GTK1. CinePaint GTK2 exists and Debian packaging work is being done by Aedan Kelly. Experimental debs are [u]here.[/u]
So to me it looks like it is all a big mess right now. But again, correct me if I am wrong.
Talking about niches… See my comment above.
I wait another 3 years and then look at it again. Meanwhile, I do some music with Logic Pro and all these Plug-Ins.
For me, the most annoying thing are the free floating windows. But the last thing I heard is that they implement an MDI for the next major version. I am glad Adobe recently changed PSE to MDI on the Mac, too.
A software for the cost of a car and then no money for a laptop?
This all seems so made up by you. How can you know that the Live CD will work on a different computer? And what if you need some more/different drivers? Again, looks made up. For me, the only use cases for Live CDs are ‘safer’ internet access, or administrative tasks like repair file systems, recover/save data and virus search.
I wonder how so many people can produce awesome results with FCP, Premiere or AVID products.
I don’t do much video editing, iMovie is fine for my needs.
Huh? You should have read my original post and name some other products other than a NLE for the cost of a new car. GIMP is debatable, Cinepaint seems to be a bad example, as mentioned above. There is no such thing as Cubase. What about DVD authoring? What about a decent video editor for the rest of us? What if you hate Blender’s UI or?
What can’t be done with PS 32 bit? And that will probably change this year with the release of CS5. Premiere will be fully 64 bit, too and come with the awesome looking Mercury Playback Engine. Of course it’ll be much less than $10,000.
Cinepaint is in the Etch repository: http://packages.debian.org/etch/cinepaint It’s in ArtistX, which is an Ubuntu-based media distro: http://www.artistx.org/site2/ I would guess that there are other Debian/Ubuntu based distros which include it. Did you check Slackware, Red Hat, OpenSUSE, Gentoo, etc.?
In regards to GTK1 vs. GTK2, I personally couldn’t care less. I am more interested in what the app does than how it looks.
Perhaps, but there is no question that these Linux programs are superior to any Windows/Mac prosumer NLE.
Open source development can be very rapid. Ardour will probably have full MIDI functionality much sooner than three years.
This is a trivial UI complaint.
I have no problem with the floating windows. In fact, I like floating windows because they allow a lot of freedom in changing the layout to suit my preferences and to suit the aspect ratio of the image.
Of course, one big window might work just as well.
It’s a matter of personal preference.
In addition, I think that GIMP 2.7-2.8 has the single-window option.
That’s a funny joke, but some people don’t like carrying laptops.
Huh? I didn’t make up anything. A lot of people use live CDs/DVDs/USBs. There are zillions of stories on forums about how people love to use them for countless different reasons.
Here is a list of live CDs: http://www.livecdlist.com/ I didn’t make up this list.
A lot of these live versions include drivers, so they work on a lot of different hardware “out of the box.”
Because they are good at what they do. The same way that anyone who is good can easily produce awesome results with Cinelerra, KDEnlive and OpenShot.
Here is a video of Andy Warhol creating art with an Amiga in 1985: http://www.youtube.com/watch?v=3oqUd8utr14
I would put my money on Mr. Warhol with an old Amiga (or with GIMP, Inkscape, Cinelerra, Ardour, etc.) against any one of the whiny UI complainers on this forum using Photoshop, FCP, Illustrator, Protools, etc.
As they say, “It’s not the equipment (or the software), it’s the person operating the equipment (or the software).
Evidently, the author of this OpenShot review has recently realized this concept: http://eugenia.gnomefiles.org/2010/01/06/less-is-more/
I’m sure that, if you are talented, that you produce fantastic results with Imovie.
If you’re fine with Imovie, than there are several OSS offerings that will do just as well or better.
Any great artist could easily finish/create fantastic work with GIMP. Same goes for Cinepaint.
In regards to Ardour vs. Cubase, Ardour should have full MIDI functionality soon. But it is fantastic in every other aspect.
If the Beatles and George Martin had to use Ardour instead of Cubase, do you think that their work would be less great?
DVD authoring? I don’t make a lot of menued DVDs. I have a feeling that the demand for DVDs is going to subside. I would guess that there are a few Linux offerings.
What if someone hates the glaring brightness of the FCP UI?
Likewise, what can’t be done with GIMP or Cinepaint?
No, it isn’t. GIMP’s UI is extremely, let me emphasize, extremely good. It was generally well thought out, it has been extensively tested and it is structured such that the user can work quickly and productively. Compared to the total junk that is photoshop… there is just no comparison.
The problem the GIMP has is that it is *unusual* and most people don’t care for that. The people who stand the most to benefit from the great GIMP UI are the people least likely to use the GIMP: professionals. I say they are least likely to use it not because it’s a bad UI, but primarily because professionals normally learn Photoshop *first*. Once you are used to that it is very hard to unlearn and relearn. Sit a pro-Photoshop guy down in front of the GIMP and he will be begging to go back to Photoshop within hours simply because it doesn’t work as he expects it to, because it is not Photoshop and that is all he knows.
The second reason professionals, those who stick it out and learn how to use the GIMPs UI to their advantage, don’t use the GIMP is an example of the same sad story we find in a lot of F/OSS projects: features. Photoshop has a lot of things it supports or can do that the GIMP can’t (yet) do. For most users these specific features don’t matter at all, but for the professional who needs them they are essential.
Criticisms of the GIMP which complain of lacking features or slow development are often legitimate. Sometimes a complaint of a lack of feature is invalid because the feature is there but it simply is called something else and was not discovered by the one who complained. Otherwise, such complaints are fine.
The people who primarily complain about the GIMP are Joe-Average-Amateur-Nobody who has his warzed copy of Photoshop, or bought it to fix red eye or add lens flare effects. These people complain about the UI, because it’s different and not familiar. They don’t use Photoshop (or the GIMP) enough to appreciate what they have.
This is you. Congratulations.
Sadly, the GIMP developers are gradually cracking under the pressure from the vocal majority of idiots who don’t know what they’re talking about. They have slowly introduced UI changes which make the GIMP more like Photoshop, sometimes at the expense of usability. If only there were another GTK image editor which would serve the needs of this class of loser! The GIMP would soldier on without such harmful interference.
The day the GIMP starts defaulting to one-big-window will be a sad day indeed.
Do they really use it *still*? The list of major movies its been used on is a bit dated, there hasn’t been a release in some time. Its interface is gimp 1.0. The “next generation” rewrite glasgow is the most absurdly unstable piece of software I’ve ever attempted to use. It is *not* getting much development from major ,movie companies. If its being updated, those changes aren’t making their way back into the trunk.
Yes. They do.
As I mentioned previously in this thread, a lot of the development probably stays in-house to give a “proprietary edge.”
Seems like conjecture on your part to me. If people who worked at ILM wouldn’t post at some VFX forums, we wouldn’t know that ILM has abandoned their former in-house compositing system CompTime for their new in-house system Zeno. Yet, you claim to know that CinePaint is still used. Any proof?
By the way, ILM recently purchased a site-wide license for The Foundry’s Nuke. A nice, proprietary compositing system that uses the FFmpeg libraries. I love Open Source!
According to CinePaint, it still has studio development (scroll down to “CinePaint Secret Developers”): http://www.cinepaint.org/team.html
There is no reason to doubt this claim, as there is no reason for the studios to abandon such a versatile, valuable tool.
Do you have any proof against this claim? Seems like you are the one making conjecture.
By the way, CinePaint is not just used for VFX and animation, it’s also used to touch-up regular footage (and in instances in which 32-bit color depth is needed).
I wasn’t aware it was used for VFX and animation. I would base my belief that its near abandoned based on the UI, and the lack of recent news,development and releases.
http://sourceforge.net/projects/cinepaint/
last file modified date of 2008-04-15
http://www.cinepaint.org/
last “News” on front page from over a year ago.
They do mention on the team page that studios develop it in secret before suddenly releasing it. So more accurately, its either being actively developed by studios secretly, or its been abandoned. No one will ever know what the fate of it is unless a new release appears from the sky.
It is not used to create effects/animation — it is used to touch-up CG work, especially when combined with live footage.
Not sure what the UI has to do with the activity of the project, except I think that the CinePaint developers are trying to make a GTK2 version.
At any rate, it is often beneficial to foster a longer attention span and to look more closely, before jumping to a conclusion.
Scroll down the Sourceforge page that you link. See all the recent bug reports? Folks are obviously busy using it.
There is even an entry by the project’s administrator from about a month ago.
If you want to get more details on these posts, click on the “Develop” link below the “CinePaint” title.
The last cvs commit, was a year ago.
http://sourceforge.net/mailarchive/forum.php?forum_name=cinepaint-c…
That does not inspire confidence in potential users.
But you are right, the main dev hasn’t stopped.
http://sourceforge.net/mailarchive/forum.php?thread_name=4B3E597F.5…
But all of the email lists are dormant for months on end before a post or two.
* The upcoming release, apparently, just isn’t being developed in a n open manner in cvs.
* Updating a web page once a year is too difficult for him.
I’m not intending to be hostile, but it took a bit of searching to discover that its only mostly dead, instead of completely dead. I wish them the best of luck.
What’s up with all this FUD?
CinePaint isn’t dead at all — it is very much alive and kicking! Read this: http://sourceforge.net/mailarchive/message.php?msg_name=4B3E597F.50…
It didn’t take me any searching to open this post — just a couple of mouse clicks. Furthermore, in the Sourceforge link that you provided earlier in this thread, I merely had to scroll down the page to find a CinePaint bug report made just 17 days ago.
CinePaint is active and growing, with a bright future.
I’m genuinely not trying to spread fud. But you can’t say with a straight face that its under publicly active development. There aren’t any cvs commits web page updates, public messages between developers. That screams dying project to me.
Compare the development to its ancestor the gimp.
* last update of website news 3 moths ago
* last source code commit 3 days ago
* 100 messages in the dev mailing list so far this month.
That is a healthy active public open source project.
Sure. Cite a page that was last updated a year ago. Do you know someone at a studio who uses it? Has he told you so? Thought so.
You know what’s big today? Integrating tools that once were separate into one big package. ILM’s Zeno can do almost everything except modelling. And look no further than Nuke which is bought by studios left and right. It has a very nice Paint module that can be used precisely for those tasks — like, for example, dust busting and wire removal — where CinePaint would’ve been used. It also has a nice optional plugin called “Furnace”, which provides advanced capabilities for automated wire removal, degraining, dirt removal, dust busting — you name it.
Today you don’t need to leave your compositor for tasks where you would have used an external paint application.
But hey, I’m probably just talking out of my rectum. And CinePaint is probably used this very moment by every VFX studio in the world. Too bad that nobody knows about it. But then, it may be pure imagination.
The Mac version of CinePaint was last updated less than a year ago, on March 26, 2009: http://mac.softpedia.com/get/Graphics/CinePaint.shtml
Now, do you think that the post production industry has completely changed in less than a year? Who do you know that has confirmed that nobody uses CinePaint?
Thought so.
I found a write-up on Zeno: “Zeno is like a collaborative scene file, a concept CG artists might know from blender…”: http://film.goeszen.com/what-is-ilms-zeno.html Looks like your proprietary software had it’s inspiration in OSS.
Which is actually what I have been saying — a lot of open code is modified by the studios for their own needs.
The studios hire more programmers than almost any other type of organization. Do you think that they are paying all those programmers to fix dust spots with off-the-shelf Photoshop?
And another thing, you seem to think CinePaint can only be used in VFX and animation shops. Well, in the film industry, there happens to be a little known style of shooting called “live action.” I know it might be hard to believe, but there is a more going on the the movie world, other than “Wall-E” and “Shrek” (although I think that CinePaint was used on Shrek).
CinePaint is used for live footage, too, and nobody is going to purchase Zeno nor Nuke, just to touch up their live action shots.
No s**t, Sherlock! Putting words in my mouth again? Nowhere have I argued that anybody is going to purchase Nuke or whatever for the sake of merely retouching live action material. I have argued that a VFX studio — I was all the while talking about VFX studios, remember? — isn’t likely to introduce an external paint application into their workflow, when any retouching job can be done from within their already existing compositing tools. There is no need for CinePaint when you already use Nuke.
Also, you were the one that started to babble about CinePaint’s bright future without any proof. Now you suddenly want *me* to prove that it isn’t used? You claim something, you prove it. From the publicly visible state I’d say the evidence is against you.
I’ve developed extensively for GIMP and I’ve used Photoshop for 15+ years. I very pro open source where it makes sense, but I also understand both applications very deeply. Claiming that GIMP is better than Photoshop is utterly silly.
Yes, there are things here and there that GIMP does better than Photoshop. Working with Animated GIFs is a good example. But those strengths are much fewer than the massive holes it has compared to Photoshop CS4.
GIMP is missing vital tools that are the bread and butter of a working photographer. This includes things like the healing brush, the patch tool, perspective image stamping, real support for actions, the quick select tool, a real brush engine, etc. The list is miles long. Trying to work as a real photographer without these tools greatly increases the amount of time spent editing.
This whole argument between GIMP and PS is out of date anyway. Most of the stuff that GIMP can handle well (color correction, cropping, basic touch ups, etc) are things that modern photographers no longer do in Photoshop. That work has all migrated to apps like Lightroom and Aperture. In fact, Photoshop is quickly becoming just a utility to photographers instead of the “main tool.” There’s nothing open source that is comparable to Lightroom at the moment.
GIMP is a great application to have around. It fits the needs of many amateurs and is a great utility for professionals as well. It’s also a great base to build your own custom image processing tools. But saying that any real professional photographer (who gets paid for their time) could use it daily INSTEAD of Photoshop is silly. It’s just not true.
That’s misleading at best. ACR uses *pieces* of dcraw. But Adobe has added thousands of hours of their own development tweaking the raw conversion algorithms. ACR is way, way better than dcraw alone.
In any case, this is where open source really excels – producing great reusable libraries of common core functions like dcraw. Where it tends to fail is producing polished, high-end applications for niche audiences. That’s no ones fault. It’s just economics.
I think it’s impossible to say OSS is “better” or “worse” than alternatives. OSS kicks ass for low level libraries and applications that apply to a wide user base. That’s why Firefox is so great. It attracts a huge number of developers to keep improving it. But that’s also why OpenShot is struggling. There just aren’t that many people in the world who want to develop a free high-end video editor with a pro-level feature set.
I wrote an article on this phenomenon for this very site half a decade ago: http://www.osnews.com/story/8146
It is interesting that, one who has extensively developed for GIMP, who is very pro open source, who understands Gimp very deeply, would claim that GIMP is missing a healing brush: http://docs.gimp.org/en/gimp-tool-heal.html
GIMP has had this function since at least version 2.4.
In regards to the other features, the patch tool is not yet included in GIMP, but the heal tool can yield the exact same results.
Also, GIMP has the Resynthesizer plug-in (which is actually more sophisticated than the Patch tool): “resyn http://www.youtube.com/watch?v=liE_sYVGFYU&feature=related From what I’ve read, GIMP had this function prior to Photoshop.
Perspective image stamping? Do you mean “perspective clone?”: http://docs.gimp.org/en/gimp-tool-perspective-clone.html Or perhaps you meant the “perspective” tool: http://docs.gimp.org/en/gimp-tool-perspective.html
In regards to the other features you mentioned, I think that most good photographers can do without them. Do you think that if Richard Avedon’s work of the 1950s, 1960s, 1970s and 1980s would be greater if he had had “real support for actions,” “the quick select tool” and a “real brush engine?”
What else is on the “list.”
Again, most of the greatest photographers never used Photoshop. I would put my money with Richard Avedon using GIMP against most others using Photoshop.
How do you know?
The main developer may not have the man power to test his software on every single distro out there. Especially Ubuntu, with its “break everything every 6 months” policy. Ubuntu is a very fast moving target. They’ve broken Xorg and gtk no less than 2 times last year. It’s not a one man job to test the software on every machine on every distros. I suggest you spend some time testing every little bit of that software on Ubuntu 9.10, isolate the bugs and write bug reports as detailed as possible. This may help a lot.
Edited 2010-01-11 09:28 UTC
Believe it or not, this is not my job to do. My “job” here is to review software, and tell the consumers what to expect from it. If there are bugs, that’s for the developers to iron out by themselves. The developer should have tested with Ubuntu, even if he might not be using it, since that distro is the most popular out there.
Of course, being the geek that I am, **I DO** bug reports. Search for me on the KDEnLive and PiTiVi bug servers. On Ubuntu servers, gnome server etc. I’m there. But I won’t be doing this forever and for everyone. “Patience has borders”, we say in my motherland.
It’s not the developer’s job either to test software. He’s probably not even paid to develop this software. He probably has another job to pay his bills and develops that on his own free time. Ubuntu might be popular, the developer might still not use it. Using the latest version of Ubuntu would be a very poor choice for a developer distro.
Nice review anyway. The advices to the developer at the end are a little unrealistic though.
Edited 2010-01-11 09:45 UTC
So if the job of testing software is not handled by the developer (you know, the person who made the software in the first place) who has some kind of clue on 1) what is expected to work or not 2) how to debug, then whose job is it? Tech journalists/bloggers?
Obligatory car analogy: do you expect car manufacturers/engineers to test their new cars, or the consumers?
Debugging a video editor needs a fair amount of intimate knowledge with it, believe me (excluding the minor design bugs or feature requests of course).
That’s the job of the testers, obviously!
Job role: software tester. Runs the software on various configurations and tries to test as many paths as possible. Then writes detailed bug reports where he explains his findings. He does not debug anything, just write reports. The devs do the debugging.
Obligatory bogus car analogy: Do you expect the engineers or the dummies to be involved in crash test?
I trust them more than the manufacturers to test the cars:
http://en.wikipedia.org/wiki/EuroNCAP
Edited 2010-01-11 15:57 UTC
I wasn’t sure if you were posting this as a tongue-in-cheek joke or not, because it’s hard to conceive that anybody would post anything so asinine — and actually believe what they’re saying. It’s just wrong on so many levels and, if this is what the devs on this project believe, then it’s no wonder that the quality is so poor.
Look, as a developer, it’s your RESPONSIBILITY to make sure that your code is as well-tested as possible before you release it. You can’t defer this responsibility to someone else. You own it. I’m not suggesting that you can’t use the help of other people. What I’m saying is that YOU own the quality of your code, and you need to take responsibility for it. You can’t blame other people for not finding your bugs. The software engineering process should include time to write solid unit tests that exercise as much of the code as possible. The unit tests should run with each and every build. It’s also advisable to create integration tests, if you need to integrate with other components. Many companies offload the job of integration to testers, but there’s no reason why developers shouldn’t be writing these tests.
Now, of course, it doesn’t hurt to have additional people testing your code, but that shouldn’t be your sole line of defense. Use private or public betas to get more code coverage. The point is to exercise as many codepaths as possible, and remove as many defects before release. This is more of a partnership between dev and test than a dividing line.
Show me a developer who says that he can’t test his or her own code, and I’ll show you a poor developer. I would never hire anyone with that kind of attitude. The best software engineers test their own code. Period.
Google doesn’t hire testers. There’s a reason for this. Developers know their code better than anyone else. Certainly, there are conflicts of interests, where a developer may not want to reveal bugs in his or her code; or, perhaps, he or she may lack sufficient vision to find their own defects. But that’s not a fundamental problem with the engineering process. It’s a problem with the INDIVIDUAL. Fix that — and you produce a solid product. Ignore it, and you produce crap.
/soapbox
Edited 2010-01-11 19:19 UTC
It’s a waste of resources to have the main developer make sure that the code works on a multitude of systems. That’s really the job of the testers.
With some of these one-man open source projects you can’t expect the developer to make sure the code works with every distro, especially if it is multimedia software.
Edited 2010-01-11 20:07 UTC
That seems… unusually non-critical, for you.
Show me a developer who says he doesn’t produce bugs and I’ll show you a liar. In any software development process, there is a developer who tests his software and there is a test team who tests the software against various gtk/xlib versions and patches. One developer can’t possibly spend 1 year testing his software, since in this one year there are 2 Ubuntu releases and I don’t count all the possible software and hardware combinations that can make the software break.
Now if you want stable software just don’t use Ubuntu. Use Debian stable, slackware, or red hat instead. At the very least, use Ubuntu LTS. If you want the latest and the bleeding edge, you are asking for testing.
Edited 2010-01-11 20:58 UTC
Complete agreement. And here here, Slackware.
How does that contradict my response? No [intelligent] developer should ever think that he or she can possibly eliminate all possible defects from his or her code. But it is possible to create unit and integration tests that are reasonably adequate to validate code coverage and conformance to requirements.
Reread for comprehension. I didn’t say that a developer had to do ALL the testing. What I said was that the developer needs to have sufficient unit and integration testing in place to validate what he has produced. That’s a different statement than saying he has to test it with every possible secondary library.
All components will only be as good as the engineering process for the entire chain of dependencies. You can produce excellent code and, if you’re depending on crappy code, you have effectively nullified any advantage of writing solid code on your side. Insist that any code you consume is properly tested; if not, refuse to use it.
The problem is that with FOSS, users will use poorly tested libraries, despite whatever you refuse to use/support. And that is the point I made. The user is free to do whatever he wants, but asking the developer to test software on anything his twisted mind can come up with is unrealistic. When the software breaks on the latest unstable Xlib, don’t complain. It was probably well tested against the proper stable build. Write a bug report.
You’ve hit upon a HUGE drawback of FOSS–or any environment that’s so chaotic. If the software is running with components that it wasn’t built against and which it doesn’t recognize, it should probably either display a warning, or refuse to run. Video editing suites tend to run as vertical stacks for a reason — it’s primarily because they’re extremely complex software that require tight integration and performance of all components in the stack. These guys should figure out how to validate all of their dependencies — stable vs test build — and lock down as much as possible. It’s a horrible position to be in.
Edited 2010-01-12 11:42 UTC
It does not work like that. Users are free. You can put a warning, and they can ignore it. You create a software that runs on GNOME, they will try to use it on Windows (cygwin, minGW and all)
It’s not the developer’s job to lock them down. It is up to the user to know what he is doing. That is what is written in the GPL (Use at your own risk)
You can call it a horrible position and chaos, it is certainly not like proprietary software, but this model has its strengths. It’s just different. It is NOT a drawback, it is a feature.
The problem is users who expect to run proprietary software that are FOSS. They’re dead wrong and they ignore it. They look for someone to blame and as with proprietary software, they’re used to blame the developer. Actually they should blame their ignorance.
Edited 2010-01-12 12:59 UTC
As I said previously, video editing places such a heavy demand for integration and performance on the entire graphics stack that the software developer should have established firm component requirements — in order to validate that the stack will function properly — and then refused to run, if those components differed. The alternative is a totally crappy experience. You can blame the user, if you like, but that’s silly. Tell the user your requirements, and then enforce them.
I understand your experience was sub optimal (and that’s an understatement), however, if you take a look at the videos posted on their web site ( http://www.openshotvideo.com/2008/04/videos.html ), you’ll see that even at version 0.82 it was functional and quite workable. Sure, those videos represent some sort of optimal experience, but the point is that it basically works and actually quite responsive on those vids.
I can’t comment on the actual [missing] functionality as video editing isn’t really my thing. I agree with your annoyance of the sparse controls, but I think it’s more of a general Gnome issue. FWIW, I don’t like it either.
I think your core-issues are mostly related to integration/testing on various distributions, and I agree it should have been properly tested on the most popular ones. That being said though, the basic stuff is there and since (as far as I understand) they include their own ffmpeg code segments, it shouldn’t depend upon a specific distro’s pre-installed multimedia frameworks. So I think the integration should be easier than on other cases that rely on pre-installed stuff.
So what do I say? give ’em some slack. Obviously a lot of work went into the project and the integration issues should be solvable IMHO. Not everyone relates to a “1.0” release the same. It probably should have been called beta, but naming it incorrectly isn’t a good enough reason for that amount of bashing IMO.
P.S.
I’m completely unrelated to this project, although I do have an interest in video in general (a mod on doom9 forums) and in OSS development (previously associated with few video-related projects and now maintain a Firefox addon).
Edited 2010-01-11 14:36 UTC
it may be very poor quality-wise at the moment. but that might be due to not enough publicity, and that’s what making an official release is for. and 1.0 draws more attention than, say, 0.1 :]
that means that at least the developer is more likely to get a lot of useful bug reports. if he gets lucky, maybe somebody will even contribute some code.
This is wrong. Very, very wrong.
I recall a time when open source developers – rightfully so – pointed out that proprietary software vendors abused version numbers to push new versions of their software down people’s throats. Higher = teh bettr.
Recently, it seems like open source developers have started doing the same thing: a project to 1.x status waaaaay before it’s ready, with a lame excuse such as “hey, this way we get attention”. KDE is of course the biggest offender.
This is not how it’s done, and heck, it’s not even needed. Check Boxee – an open source project based on XBMC. They’ve been in alpha for a long time, and yet, they had no shortage of users, and no problem getting attention. They released a beta a few days ago, and lots of major sites reported on it.
These version number shenanigans just lead to situations like this: harsh reviews that will turn people off the project. It happened to KDE, and it will happen to this project too.
Even I, with my limited knowledge of the English language, I had perfectly understood that KDE 4.0 was meant to be more of a technology preview than a production-general public release. And I tried it knowing this.
No, please, give up repeating this tale about how KDE lied to us by announcing a 4.0 release which was not complete or ready for production. NetBSD has still both KDE 3.5.10 and 4.3.1 in its pkgsrc. Guess why? They provided an older, stable alternative better suited to conservative users and production environments. I cannot recall of any new distribution using KDE 4.0 by default. Even Kubuntu let out a “remix” version of Hardy for brave users/testers, clearly separating it from the officially supported one.
So it seems pretty much everyone (I know, hyperbole) had understood what KDE 4.0 was meant to be.
I don’t GaF about the “.0” thing as long as they tell me “It’s just to say out loud that we have reached a milestone on our path”. And don’t tell me it was not clear.
Great attitude. A bit too arrogant to be a fortune teller, I’d say.
So why do you think it didn’t it happen to GNOME 2.0? Do you think it will happen to GNOME 3.0?
The whole Gnome GUI design philosophy is to surround every widget with massive amounts of whitespace!
This way they can remove customization options and unnecessary features only wanted by power users but still have the windows use as many pixels!
The lower the content/pixel ratio, the easier it’s to use!
The main toolbar of the app has perfect amount of icon size and whitespace. It’s the rest of the app’s icons that don’t, e.g. under the preview window, they are big-assed for no reason. And the tabs that say “video preview” or “timeline sequence” are unnecessary. Taking space for no reason. IMO, the developer should accept 1024 as the minimum horizontal size, and move the preview buttons on the timeline toolbar. The video tracks are too thick too.
One good thing that PiTiVi has is that you’re able to fully use it from the main menu, so even if you remove all of its toolbars (which don’t take as much space in the first place), you can still use the app fully. Plus, it has its fullscreen mode too. Overall, very usable even on a 1024×576 screen, I was surprised about that actually when I first noticed it!
Edited 2010-01-11 10:18 UTC
By this standard, vi is the easiest editor to use.
Tom
Not nearly as easy as ed.
I wonder if some of the commercial video editors for Windows can be used on WINE or worst case scenario, just use VB seamless mode and have Windows running….This (VB seamless) can also be applied on eCS for running Windows apps.
Edited 2010-01-11 10:47 UTC
Video editor in a VM? They’re not exactly light on resources.
Why not run Linux in a VM? That way you can check to see if distros like Ubuntu are still designed to be a pain in the ass for proprietary companies that make useful software like video editors.
Better proprietary app support for Ubuntu is already in the works with their integrated software store. But yeah, in general more people should accept that proprietary software is something people want to get the job done.
I doubt anything will change.
Shuttleworth needs to provide an app store so proprietary developers can charge for binaries and expect those binaries to not be broken with updates.
He doesn’t have the balls to do that. He’ll probably just add a donation button and call it a compromise.
Which, as said, is already in the works. A preliminary version was released together with Karmic.
WTH are you talking about? Ubuntu has had commercial repositories for quite some time delivering Opera, Skype and some other tools: the only missing component is the ability to pay for said software right from the installer app.
Mark Shuttleworth has said many times that Canonical has to make Ubuntu a profitable product, and making the deployment of commercial software easier for third parties seems to be an immediate goal.
I’ve haven’t seen any evidence that he is going to provide an app store for closed source. I also see no evidence that he plans on ending the 6 month upgrade-n-break release cycle.
Linking to a couple commercial repositories is not the same as an app store.
I’m sortof wondering what closed-source software they’d distribute. It’s not like there’s tons of runs-on-linux closed-source software out there, just waiting for a solid distribution channel.
I think the idea would be to provide a stable platform to attract proprietary developers. The iphone app store started small as well. It would also be good for open source developers since applications still get borked by the shared library system.
It’s primarily the customer-base that attracts commercial developers, not a particular platform. If lots of people where using Ubuntu and demanding closed-source proprietary applications, then the producers of those applications would find a way to deliver them. If there’s not a large group of potential customers using Ubuntu and demanding proprietary applications, then how appealing the platform is wonn’t really matter much, commercial applications still won’t be developed for it.
Well, at least, that’s my expectation.
Demand is only part of the equation. There’s also the cost in porting and supporting a platform and currently it takes far more work to port from Windows to Linux than Windows to OSX.
A lot of proprietary developers already use cross-platform toolkits like Qt but skip the Linux build since the whole ecosystem is designed around open source.
Predictable response: but customer demand is the larger part of the equation. The system modeled by this equation is much more sensitive to customer demand — at least, when very large numbers of customers are concerned — then it is to distribution difficulties. I won’t argue that supporting Linux would be harder, it almost certainly would be, as it’s a much more heterogeneous environment. But it’s possible, and if that’s where the customers where, commercial software distributors would figure out a way to do it reasonably economically.
Actually, right now, on the project on which I’m working, we’re using Qt, and have development against the very same code happening on a RHEL4 machine (my workstation), several Windows XP machines, and several OS X machines. The code largely works fine, unmodified, on all those platforms — minus the occasional non-fatal hiccups, I admit.
So it definitely is possible to distribute commercial software with multiple OS targets, including Linux. You can even do it with a pretty small team, with varying levels of experience and differing backgrounds.
Edited 2010-01-12 23:56 UTC
I have never denied that it is possible, it’s just a major pain in the ass compared to other platforms. I’m also not surprised that your qt code mostly worked on your Linux workstation, it’s supporting and distributing in multiple distros that leads to the real headaches.
It’s currently far easier to build and support your own distro that contains your proprietary program than it is to distribute and support your program in Ubuntu alone. Compound the fact that Ubuntu is not the only distro and Linux overall has 1% of the market and you can see why proprietary developers stick with Windows and OSX.
I was pretty sure I had read it somewhere but alas, I can’t find anything right now. I don’t know.
I guess that’s where LTS releases should come in. As with any other OS, the user is responsible of making sure that all the software they use is compatible with a newer version before upgrading. This even holds true for Windows, probably the OS with most stable API of all.
Agreed.
Edited 2010-01-11 23:36 UTC
Previewed perhaps but not scored.
Who cares what he calls it. If it isn’t done then it isn’t done.
I also had a problem with this statement:
Plus, I know very well how hard it is to write a video editor that actually works as expected (hint: more complex than developing something like Firefox).
Modern browsers are far more complex than video editors. Video files are at least standard and predictable. Web files are a god awful mess that are packed with all sorts of add-on technologies. They require incredibly complex parsers to make sense of it all and a team of developers to constantly tweak the engine.
But it does take hard work and talent to develop even a very basic video editor and I’m glad you commended him on that. I really think the “3/10” score was unneeded though. There are very few people that have the talent, drive and free time to work on a project like that and you gave him an F.
Edited 2010-01-11 11:30 UTC
Yes, Firefox would be something like a hundred times more complicated.
Would the latest Blender VSE do for you in terms of editing, keyframing, compositing (node-based), etc?
You’re stuck in a moment and you can’t get out of it… just when you make use of handy applications like video editing and audio editing, you’re really stuck with two or three so-so middle-level applications in Linux. I only have installed Linux in my computers now but sometimes the need for Windows urges, but one has to resist!
This is very frustrating indeed. I’m forced to use Windows XP at work and oh man, this OS can’t be used for even basic scripting. Even with cygwin, you have to put up with so many OS limitations, it makes simple tasks an hell to do. Very frustrating, indeed.
Uhh, yes it can. You can use VBScript and Perl, as well as PowerShell. The first one comes built-in with XP Pro (possibly also Home, I’m not sure). VB sucks, but at least you can get stuff done and the API is very powerful, almost too powerful, making simple things like copying a file require five lines of code.
Yes, you are right, there are some good scripting languages for Windows, but the problem is that the OS is not (easily) scriptable.
The API is annoying, but hardly limiting. It’s actually considerably more powerful than what you get on Unix. Check out WMI, for example, and see how much it provides. The problem, of course, is that it is a bit of a pain to use because you have to create objects, pass weird initialization strings to them, etc. But that’s a very surface problem that, for larger projects, could be handled with a convenience library.
By that logic, linux has better multimedia capabilities than Windows. You can fire up mencoder, provide the proper switches and do exactly the same thing as in Windows with $4000 software.
On a day to day basis, Windows’ scripting capabilities suck. You can’t do a ps, you can’t use grep, it is a pain to use.
I run Windows 7 on my T500 and run Ubuntu in VirtualBox. Thanks the VT-x and VB’s support for 3d, Ubuntu runs about as vast in the VM as it did on my T43 natively. So I can use Ubuntu when I need to do Linuxy things, but I can have an OS that actually works running underneath and for desktop/gaming purposes.
BTW, computationally intensive tasks should run about the same speed on a VM as natively. All ring 3 code is run directly by the CPU. Ring 0 code without VT-x is sometimes dynamically recompiled, leading to a slow down. Also, virtualizing I/O is a bottleneck, too. But if you aren’t doing I/O-bound work, VM should be as fast as native, or close enough.
you are using the wrong version.
ubuntu is not yet fully supported.
that build is old, cvs is better.
rtfm
you can’t use video editors, noob.
god uses openshot, long live god.
it’s a bug in gtk, gnome ….
upgrade ffmpeg
make up with jeff to fix preview problems
you need a config file, read the info pages
there are better ways to criticize, I can’t explain them all, I’m busy upgrading fedora using yum
how do I patch KDE2 under freebsd?
Edited 2010-01-11 13:06 UTC
Believe it or not, this is not my job to do. My “job” here is to review software…
Your job? You’re a self-proclaimed videographer… If anything, I thought you had a genuine interest in advancing the state of video on Linux, if not in all-things-video, and not write pseudo-excuses-to-yourself-so-you-can-sleep-better. You job??? **Sigh**
“But I won’t be doing this forever and for everyone. “Patience has borders”, we say in my motherland.”
While generally I like your reviews, not sure I like your personality… you sound like such a selfish prick sometimes… Most open source projects usually ask people to give something back, be it bug reports or whatever… If everyone would think the way you do, there would be no Linux…
And the way you paint yourself… First you say it’s not your job, then you show how nice you are for filling bug reports… But you not gonna be nice forever, is that right? Jeezz… Talk about egos…
Should the developer talk like that too? I’m writing this free software, but actually, I’m not gonna be nice forever you know?
Anyway, give the guy a break… He is a legend… I am too a videographer (whatever that means), and I’m too stuck with Vegas, despite hating Sony, and use Linux for all my other computing needs, but at least I can appreciate a fellow human being trying his best and giving something out for free… You’re acknowledgement of his hard work didn’t change the overall bad tone of this article, and the score of 3/10, really, should go to YOU!
Anyway, feel free to mod me… I still like reading your articles, just wish your weren’t such a one-sided rebel-toned prick sometimes…
BTW, I also don’t like your fetish for Canon products, but that’s a subject for another “rant”
If someone releases an application as a finished 1.0 version, criticism is to be expected. Being open source is no excuse for subpar quality. Eugenia’s review was very fair and detailed.
Wow look an article by Eugenia.
Hi Eugenia!
Indeed! Eugenia, we need more of you here! I miss your articles.
I too downloaded the 9.10 deps, and the 9.10 .deb.
I got it installed. I started the application… and….
The UI instantly froze. No love. Restart: No love.
This thing is a piece of trash. I can’t even run it on a Core Duo with 4gb of RAM? What the hell?
Freezes have nothing to do with strength of your processor or the amount of RAM you have.
Yeah, exactly.
The software sucks.
Your review had a tone of being upset – that is totally unjustified. This is code shared with you for free. Be grateful. Your degree of gratefulness can be determined by the quality of the code. I think you should feel free to offer constructive criticism, and some of the concerns you raised are valid, but bring that up from a heart of gratitude, not of indignation. Being mad at this generous contribution to FOSS proves you just don’t get it, and would be better off holding your tongue.
Here’s a hint for you. A *user* doesn’t give a flying f–k about the code or the quality of it. He doesn’t give a damn about all the religious, moralistic, philosophical overtones of Open Source. What he actually cares about is if this software gets the job done or not. It is a downright ludicrous idea to give software leeway or even a free pass when reviewing it, just because it happens to be Open Source software. If you don’t want your software to be judged, don’t release it. It’s that simple.
I am not saying you have to like the program, just respect the fact that somebody is donating their work and at least giving you the chance to benefit from it. It isn’t about a religious devotion to FOSS, or giving bad software a free pass. It’s just common courtesy. You don’t spit on someone who is being generous. If it is no good, or of no use to you, fine, say so, hopefully with good reasons, but don’t be angry and belittle them. Constructive criticism is an essential part of open source development, but the attitude in this review was just mean. So much so that I feel it largely negated any of the legitimate issues the author touched upon.
If you paid for a service or product it makes sense be upset if reasonable expectations aren’t meant. Not so with free stuff. I feel like the writers for a site like this should get that.
And that’s where I wholeheartedly disagree. It’s one of the biggest fallacies Open Source has to offer. Why should you not have the right to be upset if free software doesn’t meet your reasonable expectations?
You forget one thing: there’s no myriad of commercial consumer-grade applications available on Linux, as an alternative to free software. A user can’t say: “Okay, this free video editor doesn’t cut it, I’m gonna buy Adobe Premiere.” (Let’s forget about Wine, it’s a crutch, an ugly, ugly hack.) He has to make do with the free software that’s available to him.
And every damned Switch-to-Linux-and-you-will-be-happy-ever-after website out there promises the same: “Linux has not only thousands of applications that are free, no, they also are equal to or better than their commercial counterparts. And did I mention that they’re free?”
And you’re telling me that someone who has switched to Linux, has no right to be upset about the sorry state of many free applications in particular genres, simply based on their free-ness? Linux doesn’t need video editor No. 100. For the time being, it just needs one that works well, and can at least rival last year’s iMovie or Premiere Elements.
Free-ness isn’t a criterion for quality.
Sure. I think the big issue here is that for over a decade thousand and one lemur2s have been telling us that Linux and open source is the greatest thing ever. But you can not have it both ways.
You can not have it in the 1990s way where most of the users were developers and the 2010s ways where people are trying to push Linux and open source for the common consumers. The latter group has all right to bitch and moan.
Exactly.
I agree. Though I see a difference here too; I wouldn’t go and flame say the Haiku community for the flaws in their system simply because that is predominantly a project done by volunteers or at least certainly something that is not pushed by the fanboys as a ultimate solution.
Yes. Not a necessary nor a sufficient condition.
Agreed. Nobody should ever blame or criticize users for finding defects in software. As for features, software can always be improved.
Suppose you were a new programmer and thought you would bless the Linux – opensource community with your time, skills and hard work. Okay it is not perfect but it works and you are proud of it. Then along comes Eugenia Loli-Queru. Wow, forget it. if this is how I’m going to be treated for trying to help. Please try something like “Good job but it would be nice to see this or that”
That is not how you motivate someone to do better, that is how you get them to give up all together.
Jonathan Thomas please know that we do not all fill that way. Keep up the good work.
When you produce something — regardless of whether you give it away or charge for it — you should expect some degree of criticism. Eugenia isn’t criticizing the developer, personally. She focused exclusively on the software and its shortcomings; frankly, the worst criticism is the fact that it crashes once a minute. Everything else is just noise, because it is clearly unusable in its current form. The developer should just take that criticism humbly, learn from it, and work harder. As Edison used to say, “genius is 1% inspiration and 99% perspiration.”
You missed my point, “Encourage rather then discourage”. It is how it was said. For instance you could say “When I look at you time stands still” or you could have said “You have a face that could stop a clock”. They both say the same thing. She/He could have said the same criticisms to the developer and also inspired him not to give up and do better. That is why I titled this comment “Does this guy work for Micro Soft”
Sorry, but I can understand the difficulty with encouraging an app developer whose crashes once per minute of use. I just think the dev made extremely ill-advised engineering decisions, and is paying the price for it. Never blame the user.
and all anyone can do is complain. I have been waiting since 2006 to edit my wedding video in linux (having exhausted all of the other video editing options) and can finally say that I’ve done it with OpenShot. It’s not perfect, but can anyone here say that this isn’t the best video editor for linux?
It has developed at such a rapid pace, and all by one person, that I can’t believe this article wasn’t skewed in the positive a little. Oh wait, it was Eugenia that wrote it. She loves to hate Linux video editors. She is perfectly entitled to her thoughts, but I can’t help but get an extremely negative spin from her.
The very first time I used OpenShot, I emailed John and thanked him for doing a great service to the free software world. I suggest anyone that likes OpenShot to do the same. I can’t wait to see where the project goes from here!
Edited 2010-01-12 01:12 UTC
There was no “negative spin” in my review. It was objective, 100%. If anything, I’m LONGING for a stable video editor, for years now! My husband was sitting next to me, researching some geohashing stuff while I was testing the app. He could see the freezes too, he could see the crashes, he could see me dropping to a text screen in order to kill the frozen X11. So no. I didn’t dream up the instability, neither I live in a different quantum dimension where version 1.0 had not been released yet. What I wrote in the review, is 100% truth. No negative spin, no under-the-table sentences or meanings. For those who know me from the olden OSNews days, or even just read my blog, they know well that I’m a straight-up person, even if they might not always agree with me.
And the truth is, the thing is super-unstable. It’s unusable. This project is definitely not “the first good video editor for linux”, as you wrote in your title. Don’t believe the hype. So I would appreciate it if you’re a bit more careful with your accusations.
Eugenia’s reviews tend to be pretty blunt and to the point with some personal opinions thrown in a couple places, which can rub some people the wrong way. But I don’t see the problem, she’s just being honest.
She might be wrong about a few things (video editor more complex than a browser engine???), but she gives a pretty good account of what an end-user will experience with with this, which is the whole point of the review. So stop attacking her for just reporting what she saw.
Common misconception.
First and foremost, OpenShot uses MLT, which is basically the KDEnlive engine/backend (in crude terms, bear with me here). MLT/KDEnlive have existed, if I am correct, since 2002. OpenShot seems to be “just” piggybacking on the KDEnlive project. Writing a frontend for video editing in a year is possible, but writing a complete frontend + engine/framework/backend in a year? No way in hell.
I can’t imagine how deeply annoying it must be to other video editing projects to see OpenShot arrive on the block, reuse the engine of an existing project and basically clone it in PyGTK “cause KDE/QT stinks and stains my pure GNOME environment” (note: I actually a GNOME user/fan, and this quote is a satyre of the mentality I see around the project’s users) and then subtly let users believe (through hype and insufficient clarifications) that the whole thing was written from scratch in a year. Urgh.
Second point: while Jonathan is the primary figure on the project, as far as I know (I could be wrong), he is far from being alone working on the project.
I am a cinelerra user for most video. The CV branch is my favorite videa editor when compiled on a stable environement (I have to use a VM, because it is kind of hard to keep it sane enough that cinelerra stay open for more than a minute). The features are there and work, including more advanced like chroma (blue screens).
Jashaka gave me a lot of hope some years ago, but that POS was a mess and I never really understood how it work.
Kino is great for quick DV editing. It is by far easier than Windows Movie Maker and as easier as iMovie. It also have some transitions and plugins, but the only compatible format is DV and it is too linear, you can trim individual clips and change order, but that’s pretty much it.
KDenLive have never been stable for me, until I tested it again 3 days ago, and it worked (for once… I follow that project since 5 years). The editing was easy and the interface is similar to OpenShot/WMM/iMovie 03-06 and most “consumer” video editor. It lack the powerfull feature of Cinelerra, but the interface is less terrible. It is flexible, you can ajust the UI as you like and after all, it work!
However, Cinelerra is still by far my favorite.