Linux on desktop computers will begin taking off in mainstream markets in the next 12 to 18 months, Novell President Ron Hovsepian has predicted. Linux has been widely used on networked computers called servers, but it has comparatively little success on personal computers, beyond technically savvy users. Many companies have argued the open-source operating system is on the verge of breaking out in PCs and have been proven wrong. But Hovsepian sees some changes that he believes make the market ripe.
If Linux really does take a big bite (say 30/40%) out of the windows market then the one thing I think that needs a some work is productive crossplatform GUI development environments. As a developer I have not seen anything on Linux or Mac to match the tools we have on Windows (and not just by MS either). Yes people will shout eclipse at me but apart from a few areas it excels at (eg refactoring) I think it is behind in other significant areas.
At the moment we have probably two viable platforms for the develpment of *large scale commercial apps* (so not python etc), Java and QT. Given Java’s continuing problems on the GUI front this isn’t likely to go down well with end users (I had to use a Java app this morning and it’s GUI was pretty bad). Qt however is ok, although the QT apps I’ve used tend to look a bit dull in this modern age of eye candy.
The one great and very significant advantage that the windows era brought was a single platform to target, with all the benefits of economies of scale and consistency (both to the user and developer). If those days are about to end then we should avoid increasing our maintenance costs (and hence end user costs) by having multiple platforms to target. I guess if you live in the open-srurce world where economics don’t apparently apply then costs are not an issue.
> Given Java’s continuing problems on the GUI front
> this isn’t likely to go down well with end users
> (I had to use a Java app this morning and it’s GUI
> was pretty bad).
Java does’t really have any problems on the GUI front end, except for the cross platform “metal” look and feel. But you don’t have to use it. Check out some of the JGoodies Look and Feels. They make both Gtk and Qt look dated and ugly. JGoodies has some very elegant themes. And you can just bundle the jar with your app so the end user doesn’t have to do any work to make the app look nice.
try REALbasic its cross platform for Windows, Mac and Linux.
http://www.realbasic.com
Java does’t really have any problems on the GUI front end, except for the cross platform “metal” look and feel. But you don’t have to use it.
If you are talking about the present situation I wouldn’t agree with you. In Java 5 you can’t even fully use the keyboard on Linux unless you run US locale. The theme support could see some improvements, the responsivenes could be better.
The new Java 6 will change some of this, among other things the over seven year old keyboard bug is allready fixed in the current beta, and it feels a lot snappier. Still I think Java on Linux have a long way to go before we can count on it as a cross platform development tool for the desktop.
There seam to be a bit of a shift in direction for Sun. It seams that they actually intend to have something to work on the desktop this time, but as they neglected it for so long, it will take some time before they catch up. Java 7 perhaps. Till then, the best we can do if we need to be cross platform is probably using QT.
> Till then, the best we can do if we need to be cross platform is
> probably using QT.
Qt has its own problens though–the biggest one being its licensing requirements. The extremely high cost of a cross platform Qt license effectively locks out smaller shops who are just trying to get started. And the alternative (licesning all your stuff under GPL) just isn’t feasable sometimes.
Qt has its own problens though–the biggest one being its licensing requirements.
I would love everything to be free, but unfortunately it just isn’t going to happen. Good development tools require time and investment, and that money has to come from somewhere.
The extremely high cost of a cross platform Qt license effectively locks out smaller shops who are just trying to get started.
Ahhh, the mythical small development shops. The vast majority of small development companies develop for Windows, and they spend a hell of a lot more on development tools and software than a handful of Qt licenses.
> The vast majority of small development
> companies develop for Windows, and they spend a
> hell of a lot more on development tools and
> software than a handful of Qt licenses.
Actually, the vast majority of small development shops spend little or nothing on development tools cause they use open open source tools. And the ones that do spend money? Well, Visual Studio.NET Enterprise edition is less than 1/6th the cost of a cross platform Qt license.
So no, the vast majority of small shops do not spend anywhere near the amount of money that Qt costs on development tools.
Actually, the vast majority of small development shops spend little or nothing on development tools cause they use open open source tools.
Do you have any figures for this hyperbole?
Can you do everyone a favour and go and work for a development company, or go and do some research, before posting utter crap that you’ve read on some GTK forum somewhere?
Well, Visual Studio.NET Enterprise edition…
Since we’re talking about non-Windows development, .Net is a cross platform development framework that works on non-Windows platforms is it? No, it isn’t (sorry, Mono doesn’t count – .Net is not a cross-platform framework).
Development companies also spend money on an awful lot more then Visual Studio .Net. A majority use MSDN and that kind of developer network support is included in your Qt purchase, and then there’s the peripheral costs of other Microsoft software you need. Can’t do development without them. Many also use commercial Java development tools and other add-ons as well.
You would be wise to do some research on just how much it does cost to do development in any company.
And of course, we’re talking about upfront costs here rather than other costs (increased developer time) and the long-term savings you get from Qt versus Microsoft .Net, cross platform development and woefully inadequate (certainly cross-platform) development tools like GTK.
So no, the vast majority of small shops do not spend anywhere near the amount of money that Qt costs on development tools.
Very wrong.
All this is beside the point though. People want good development tools if they ever want to develop for a Linux platform (as do open source developers), and the crop of free stuff around, as an overall answer, just doesn’t cut it.
Edited 2006-04-12 14:18
> Can you do everyone a favour and go and work for
> a development company
I’m a software engineer at a development company, thank you very much. I suspect you are the one who does not work for a development company if you don’t realize how tight budgets are for software development these days.
> You would be wise to do some research on just
> how much it does cost to do development in any
> company.
I already know how much it costs. A lot. But budgets are still tight.
> Very wrong.
I’m not at all wrong. I’ve been a software engineer for 20 years, and I know how the economics work. Budgets are very tight.
Edited 2006-04-12 14:55
> Qt however is ok, although the QT apps I’ve used tend to look a bit dull in this modern age
> of eye candy.
Take a look at Photoshop Elements (not regular Photoshop) – it’s built upon Qt and I don’t think it looks dull at all. While it’s only available for Windows and OS X, it should be portable to Linux etc. if Adobe wanted to.
Since Adobe has been seriously keeping an eye on Linux, I wonder if the use of Qt is intentional, as they may plan to port some of their products natively to Linux to dip their toes in the market. If Adobe came onboard, this would open a few doors for Linux.
http://news.com.com/Adobe+dipping+toes+into+desktop+Linux+waters/21…
Take a look at Photoshop Elements (not regular Photoshop) – it’s built upon Qt and I don’t think it looks dull at all. While it’s only available for Windows and OS X, it should be portable to Linux etc. if Adobe wanted to.
I wouldn’t hold my breath for that; Adobe was getting a beta version of Framemaker tested on Linux, then for no reason at all, they pulled it off the market.
Preasure from Microsoft? nope; more likely ‘can’t be bothered given the size of the user base’.
Yeah sure, a few million potential customers cannot bother them
Yes, millions of customers who stay quiet rather than getting out and causing a fuss.
What does this inaction tell to Adobe? no one in the Linux community gives a toss about ISV’s and software developed and sold by proprietary software establishments.
“What does this inaction tell to Adobe? no one in the Linux community gives a toss about ISV’s and software developed and sold by proprietary software establishments.”
They may well be finding themselves with a decreasing window of opportunity as well. Either they get in early enough or they may find an adequate alternate is coded and popular before they have the opportunity to repeat their windows success. So if they resist for too long they may find themselves wrongfooted in the future.
Linux users will pay for apps, I’ve got a shelf full of commercial linux software I’ve bought, but I’ll use FOSS apps unless the commercial software has a tangible benefit. The longer they wait the less obvious that benefit becomes, as alternatives spring up to forfill their missing functions.
I don’t dual boot, I’m very happy with the Linux desktop (my YOTLD was a few years back). so if you want to sell software to me it needs to be for my OS of choice, otherwise no sale. It’s really as simple as that, I can just spend my money on other things.
They may well be finding themselves with a decreasing window of opportunity as well. Either they get in early enough or they may find an adequate alternate is coded and popular before they have the opportunity to repeat their windows success. So if they resist for too long they may find themselves wrongfooted in the future.
How are they going to ‘find themselves with a decreasing window of opportunity’ considering not one opensource application can touch the likes of Microsoft Office, Photoshop, Indesign, Dreamweaver, Quicken, 4D and MYOB, to name a few, in the depth and breath of functionality and ease of use.
Sorry, unless you’re some sort of masochist who loves pain, missery and low productivity, then sure, jump on your bike, and enjoy the world of half-baked and poorly opensource application projects, managed by people who, if you dare offer advice (like I did with GIMP), you’re abused to buggery, and told to shut up and put up with it.
Sorry, the opensource community talk about free love, freedom and free code, but what it is more like is a geek feast of coders with chips on both shoulders, with arrogance that exceeds the average French man.
How are they going to ‘find themselves with a decreasing window of opportunity’ considering not one opensource application can touch the likes of…
That’s probably why he wrote “decreasing” not “closed”
And it does refer to the market not to open source competitors, a closed source competitor could also take the chance to secure a wide portion of the new market before any of the “usual suspects” show up
Take a look at Photoshop Elements (not regular Photoshop) – it’s built upon Qt and I don’t think it looks dull at all.
Actually, it’s Photoshop Album that uses Qt.
And I think recent versions of Qt use the native theming APIs of Mac OS X and Windows XP to draw their widgets, so the look should be pretty good, even if the layout and behavior is not spot on.
Well Novel is pretty heavily invested in Mono as a Development platform I believe. With a little more work MonoDevelop could become a very good solution to many of the great Windows IDEs. In fact, the most recent Mono has even introduced a drag and drop GUI builder that uses GTK#.
A bit more polish from Novell and Mono could be to Linux what .NET is to Windows.
At the moment we have probably two viable platforms for the develpment of *large scale commercial apps* (so not python etc), Java and QT. (…) At the moment we have probably two viable platforms for the develpment of *large scale commercial apps* (so not python etc), Java and QT.
Don’t forget wxWidgets – though that might be even duller than QT
.
“Qt however is ok, although the QT apps I’ve used tend to look a bit dull in this modern age of eye candy.”
Compared to what? Windows? Either you have used some unusually ugly programs or you are just plain wrong.
Why not python? If you’re thinking about the source being available, python code can be compiled just like java.
And what is wrong with QT’s look – after all, it’s skinnable.
It’s tough to get excited when someone says that every year. I’m not saying it’s not going to take off, but geez. Heh, one could set up a cron job for once a year to make a pronouncement about Linux desktops taking off.
0 15 4 10 * /usr/local/bin/linuxdesktop.sh
That was seriously funny. Even the resident Linux-fanboy here laughed at it.
🙂
Too bad OSNews doesn’t let you mod people funny, so I just modded you up
You’re right – every year is touted to be the linux desktop. But so what? One of these years will be, we just don’t know which one yet =)
Like some wise person before me said: “Everything has once happened for the first time…”
And to be honest – Linux IS growing exponentially (although at very low points right now), and it’s constantly moving forward technology-wise.
I think the mainstreaming of Linux has gone largely undetected, likely because it has happened in a way people didn’t expect. Linux has already basically entered the mainstream. It runs on everything from Tivos to Sony WEGA TVs to internet routers. When the PS3 comes out, a copy of Linux will be on every one of them. Whether you realize it or not, there is a pretty good chance that you already own a machine running Linux. Beyond that, Linux has taken off in lots of desktop markets as well. In the movie industry, many arists have a Linux box on their desk. The same thing is true for the scientific/engineering worlds. Outside of the United States, in places where the cost of software licenses is a bigger issue, Linux has been implemented on lots of government and educational machines. Linux has stealthily snuck its way onto lots of machines that “regular people” use every day.
Now, if you’re expecting a day to come when 30% of grandmas in the United States run Linux on their home desktops, you’re probably in for a dissapointment. The market is just not set of for that. Apple is a multi-billion dollar company, with enormous mindshare and a huge marketing budget, yet cannot break 5% marketshare. At the same time, its important to realize that grandmas in the US is a fairly specific market segment. Home users are a large market, to be sure, but its foolish to act as if its the only market. Linux can be very successful “in the mainstream”, without ever capturing a substantial share of the home user desktop.
Yeah I admit that someone does say this every year.
But
Novell has been dumpin a lot more money into making the kind of cool things (Xgl, F-Spot, Beagle, etc) that users want. I’l believe this sort of claim more from Novell, who appears to be backing it up with cold hard $$$, than I have in the past.
It’s tough to get excited when someone says that every year. I’m not saying it’s not going to take off, but geez. Heh, one could set up a cron job for once a year to make a pronouncement about Linux desktops taking off.
0 15 4 10 * /usr/local/bin/linuxdesktop.sh
Very funny, well done. Unfortunately, I’d have a hard time modding you funny if I could because it’s not just funny but completely true. It’s also been the same people using this cron job as well, since it’s always been someone at Novell or Ximian (not just them, but especially them) before them that’s come out with this. It happened in 2000, 2001, 2002, 2003, 2004, 2005 and now 2006. The problem is that no one is saying anything different when they do it. “Desktop Linux will take off over the next eighteen months…”, “People who don’t need a full-featured desktop will go for it…” etc. without any explanation as to what will actually motivate people to go for it and years of evidence to the contrary.
If there was something in what Hovsepian was saying that was different to what’s gone before then fine, I couldn’t be happier and I’d look forward to the next couple of years. But there isn’t. People pushing desktop Linux don’t seem to realise that people are not going to buy into it just because they have a handful of people who apparently don’t use all the features of their Microsoft systems. Hell, that’s most people.
All of these people just don’t comprehend that there are some pretty severe hurdles to be removed before a non-Windows and Microsoft desktop becomes viable. For example, they need to be able to give away their desktop for free (which puts the skids on SLED and Novell’s revenue model :-)) to create a ready made userbase and market in the same way that Firefox has done. In order to do this they need to make the desktop good enough by providing top-notch GUI tools (which means using a proper development platform), make people aware of the development framework to be used (whether they pay licenses for it or not etc.) and create a sane way to install software for both users and developers. It also needs to be easily fully updateable. They then need to interweave this fully into the desktop.
Then on top of this, the roadblocks that keep people using Windows even when an alternative is available need to be removed. People always complain that they need support for the doc format or for Windows Media. After the above has been done, and a reasonable userbase has been established (or can feasibly be established), there needs to be an easy (one-click if possible) process to convert all a user’s Microsoft Office files to ODF. Support for Windows Media needs to be negated by using an open alternative. Ogg Vorbis is probably a good option, and a solid streaming media server needs to be brought in or developed, along with the right plugins, to supplant Windows Media and hopefully other proprietary formats.
The good thing about getting what’s in the first paragraph done is that once those pieces are in place, and a userbase starts to grow, third-party developers (those developers that people talk about but don’t currently exist) will step in to create tools that will make the whole process even easier. However, it’s getting there, and using the right technology, that’s the problem.
I like linux and love suse.. But this is getting old. Maybe if they just made the software and shut up instead of predicting the beginning of a new paradigm it would be here already…. Talk about it when its out.. Dont follow Microsoft’s lead with “vaporware”. (not that its really vaporware,, but you get the point).
I hope the guy is right. MS really needs some competition for the beast of an OS that Vista is.
time for the year of the lunax dekstop1111 articles to get into full speed
Instead of doing the work to make a consumer-ready OS, they just serve up a half baked mish-mash of fragmented community developed shovelware and then make a press release claiming success is ‘just around the corner’.
This is not the way to make a polished, consistent and reliable product.
I’ve been a happy Linux user (I actually enjoy spending countless hours working out how to beat my OS and software into submission) for many years now, but I can’t say any of the distros are complete and polished enough for me to recommend to a non-technical user, and I doubt SuSE’s latest incarnation will change that at all.
I don’t see Linux succeeding on the desktop for a long time.
I can’t agree – Linux desktop is perfectly ready for the non-technical user. I watch what my girlfriend does with her Windows PC, and every task is completable in a similar manner with Linux.
If her Windows PC has a problem, I’m asked to fix it. If she used Linux, I’d be asked to fix that too. She rarely, if ever, installs new programs (and even then, only at my suggestion for a task she wishes to complete) so she has no “favourite software” that is not usable in Linux.
It is the “mid-range” user – ie, with some computer knowledge they’ve built up over time, who uses a wider variety of Windows software – who Linux is not ready for.
Edited 2006-04-10 23:20
If her Windows PC has a problem, I’m asked to fix it. If she used Linux, I’d be asked to fix that too. She rarely, if ever, installs new programs (and even then, only at my suggestion for a task she wishes to complete) so she has no “favourite software” that is not usable in Linux.
I’ll disagree with you there; that scenario you gave sounds alot like my mother and father (father used to work for Burroughs before it was bought out my Sperry, and merged to create UNISYS).
They have applications they like; my mother loves using Publisher; she knows how things are laid out, she has written down instructions on how to get set tasks done – is she suddenly going to throw out all that knowledge, in her case, took time and patience to learn, simply to get the fuzzy feel good factor of going to Linux (or any other operating system)? hardly.
My father, is a Microsoft Office whore, thats the only way to describe him; he has been using it for years, and before that he was using a combinaton of Harvard Graphics, Lotus 1-2-3 and Wordperfect. For him, wine *could* solve the problem, but that is still buggy, and not everything works nicely.
Now, this isn’t an attack on you, but I think the first thing that should be concerntrated on is getting wine up and running; atleast then, worse case scenario of companies not porting their applications accross, end users can still run their favourite applications, be it under a translation layer.
Well, my father is running Gentoo (with my brother as maintainer of the system, when I’m not around).
Linux is perfectly ready for mainstream use. The issues mainstream users cannot fix on Linux are identical to the issues they cannot fix on Windows.
Wine is btw. working quite fine, though the installation process could be made easier.
But apart from running IE to access one particular FF-unfriendly site, I don’t use Wine much.
I consider mono to be much more important.
Wine is btw. working quite fine, though the installation process could be made easier.
True, and there are some iffy behaviour; for example, with printing, but like I said, it works well; the problem is, however, my mother/father share one computer, so if there is going to be a migration, it’ll have to occur when both can move.
I haven’t come accross many IE only sites, if a site doesn’t work with Firefox, I assume they are unimportant, and I don’t visit it again.
As for mono; I think that Novell is holding out, hoping that when .NET does take off (when ever it actually does), Novell is hoping that they can grab onto the tail end of it, and see some of those ‘.net native applications’ being made available on Linux in the way of Mono.
With that being said, however, migration of applications from Windows to UNIX isn’t as hard as some people try to make out; hell, Mainsoft has made a scoop of profits off selling a product that is directly derived from Microsoft Windows source code – so the tools are there, its the software companies who are choosing not to port.
Want to change their minds; support hardware companies that provide good support (Nvidia) in the way of binary drivers, and simply lobby companies through mass movement to get the likes of Adobe to port their products over.
I mean, if a bunch of political hot heads can get together to put together an anti-Kerry or anti-Bush lobby group, I’m sure a group of geeks and enthusiasts can get together to create some noise which the ISV’s can’t avoid.
Edited 2006-04-11 06:03
haha “GENTOO is so easy my DAD uses it!”
Well, it’s easy to use. I made it clear that he didn’t maintain it.
It’s usually wise to actually read all of the post before replying
If her Windows PC has a problem, I’m asked to fix it. If she used Linux, I’d be asked to fix that too.
True, ditto here. The question is: are there enough of you to go around? There’s a whole eco-system out there that can support Windows (some better than others :-), but Linux is seriously different.
Similar reasoning applies to the corp desktop: Dell et al. have oodles of support people that, at least in theory, can support Windows desktops ( http://www.inktank.com/AT/index.cfm?nav=390 ), but getting enough people well-versed enough in Linux *and* capable of talking to customers takes time.
It’s not so much a question of “Linux is ready for the desktop” but more of “the environment is ready for Linux”. And of course, if you state the first often enough, the latter will eventually become true.
Yup, sure.
I dual-boot SuSE 9.2 and XP – it originally took me 2 weeks of tinkering before I had SuSE running the way I wanted. Then I did an Update which borked the install – next time was quicker, only one week.
Last week someone asked me to fix their PC with a borked XP – no CD of course. Says I, “Try Linux!” and they agreed. Gigabyte mobo, Athlon 800 CPU – bog standard. Put on Mepis – stuck in 640×480 despite a whole day on the Web experimenting with XF86Config-4. Dunno what would have happened with sound. Update – half the .deb repositories are out – fart around some more and then screw that – go to Ubuntu, though I’ve never liked Gnome.
Ubuntu, vile turd-brown default theme – try to change that. Loads of obscure menus designed by a retard – finally succeed. Finds serial modem, but refuses to use it. Another fruitless day on the web. Piss on that – dig out old Win98SE disk
45 minutes later working install. Load up the unoffical updates, OOo 2.02, Firefox, Thunderbird, AVG Free, Agnitum free firewall and DeepBurner – another hour.
Off ye go. Nice crisp little system ready to roll. OK, so I did a bit of fine-tuning and set up some maintenance cronjobs but I didn”t have to.
Lets be clear, I really dislike MSs attitude and business practices – enough to avoid them whenever possible. I love Linux and the freedom that it gives me.
But “ready for the desktop”? No way José… MS has spent hundreds of thousands of man-hours studying usability and it shows.
With Windows Vista coming out in all flavors even cheaper ones and Windows XP is around and will get less expensive I don’t see what room is left for the linux desktop.
People have to have a need or a reason to switch. If Windows Vista is what Microsoft says it is (ie. Much more secure and much more stable) and has plenty of polish and features and runs your software that you have, then I don’t think people are going to switch.
Why would I want to switch if I am happy with Windows XP or Windows Vista?
Windows Vista or Windows XP can run MONAD which is a nice shell with .net 2.0 abilities to do hardcore stuff, why would I need anything else?
I mean even a lot of the Linux applications are going to be running on Windows now. So what exactly is the point?
I think Novell is going to be disapointed when 18 months comes up and people are still on Windows XP and Windows Vista.
Everybody has different reasons. Personally, I got tired of Windows controlling the way I work.
You don’t sound like you have used Linux for more than a day or two. That means that you would have never gotten used to some of the things that make Linux a superior environment for getting things done. For example, it’s like tabbed browsing. If you’ve never seen them and never used them long enough to get hooked on them, you can’t understand how nice they are. Same thing with rocker mouse gestures. To appreciate Linux, you have to *USE* Linux for awhile. You can’t just compare features and say, “Windows does that, too.”
You ask, “Why?” For me, Linux is just more comfortable.
Oh. And, about Monad, I haven’t tried it. But, you mention it as though it makes life easier in Windows, like Bash does for Linux. That alone is a testament that Linux is a better environment for some things. I do have to use Windows at work. And, I thought about trying Monad, is it just a scripting language, or does it give you true Shell shortcuts, like BASH.
With Windows Vista coming out in all flavors even cheaper ones and Windows XP is around and will get less expensive I don’t see what room is left for the linux desktop.
What planet are you on ? Microsoft has NEVER reduced the cost of an OS, even if newer ones get released.
I mean even a lot of the Linux applications are going to be running on Windows now. So what exactly is the point?
So that you have the performance and stability of Linux running underneath your applications,
I think Novell is going to be disapointed when 18 months comes up and people are still on Windows XP and Windows Vista.
Novell might be disappointed, but they wont be half as disappointed as Microsoft will be. They will be releasing Vista for hardly anyone to upgrade to it, plus they will loose existing customers to Linux and Mac OSX.
Bad time indeed for Windows and its supporters
Here’s what Novell really needs to do if they want the home desktop market:
1.Give us dvd. Sure it’s patented, and therefore you need a licence. Just. Pay. For. It. Already. A licence for dvd-video is $2, and they’re charging $60 for the boxed copy of SuSE, give me a break.
2. Push Autopackage. Nobody seems to be supporting Autopackage heavily. Novell should divert funds toward porting native apps to Autopackage. There are tens of thousands of useful Linux apps out there, but few are available to newbies. One of my best friends returned to Windows due to this pesky problem.
3.Push for the porting of important Windows software. TurboTax, Quicken, etc. Thankfully, it seems that’s what they’re planning to do
http://www.novell.com/coolsolutions/tip/16646.html
My father decided it was time for him to upgrade from his antique WinNT box. I managed to convince him that Linux was the ideal choice. He’s an old DOS nostalgic, and was quite interested in the idea of having a useful text prompt to play with again, not to mention the money saved. “Does it run TurboTax?” One would be convert gone. Many people are compelled to stay because of a single app, their own personal ball-and-chain. Even if Novell offers to port an app themselves, at a loss, the net gains from new users buying boxed copies may well be worth it. Not to mention finally breaking the chicken-and-the-egg problem
Defcon,
Re: “1.Give us dvd. Sure it’s patented, and therefore you need a licence. Just. Pay. For. It. Already. A licence for dvd-video is $2, and they’re charging $60 for the boxed copy of SuSE, give me a break.”
Microsoft doesn’t include the ability to play encrypted DVD with Windows XP so why does it surprize you that Linux distributions don’t provide this with a default installation? Anyway, libdvdcss can be obtained for free for encrypted DVD playback though Linux developers are in question regarding it’s legality and reason why it’s not packaged with the included video players.
Re: “2. Push Autopackage. Nobody seems to be supporting Autopackage heavily. Novell should divert funds toward porting native apps to Autopackage. There are tens of thousands of useful Linux apps out there, but few are available to newbies. One of my best friends returned to Windows due to this pesky problem.”
If you’re friend had actually done research on the various Linux distributions as well software both commercial and open source available he/she would of noticed differences. Binary packages used are similar to those used for Windows. RPM which is used by popular distributions such as those provided by Novell, Red Hat, Mandriva, etc is a standard that’s been used for several years. Where as Autopackage is a fairly new approach that has not even reached final stage in developement and not recommended by distribution developers, including those that are Debian based. Anyway, it’s obvious your friend as well yourself are not aware that there are over a thousand open source applications compiled for RPM and there’s also commercial software compiled for RPM (ie: Maya, XSI, Shake, Smoke, Cedega, Linux games such as Quake 4, etc).
Re: “3.Push for the porting of important Windows software. TurboTax, Quicken, etc. Thankfully, it seems that’s what they’re planning to do.”
While I do agree with wanting more commercial applications ported to Linux you have to understand developers don’t really listen to OS developers but instead to consumers who may purchase their software. So if you want a particular product ported then contact the developer of the product you’re interested in and request a port to Linux. Several people have done this for Pixologic’s ZBrush which is currently under discussion for a possible Linux port. It really comes down to consumer demand for a port or feature request. If there is enough interest in the port to Linux a developer will listen to their customers.
libdvdcss is perfectly legal in most countries, with the possible exception of USA.
Windows installers are similar to linux model, but with one big difference: You have your application installed to separate directory, possibly even in under home directory. If you need newer version of library for app, you put it’s binary in application directory and app will use that one. If linux app requires, for example, new version of Qt, you are stuck (without hacking throughout the system).
WIndows model also can enable keeping of multiple versions of apps installed, while under linux model they would probably be overwritten.
I think that the best thing about autopackage is local-user install functionallity, for system install it is better to use debs/rpms.
WIndows model also can enable keeping of multiple versions of apps installed, while under linux model they would probably be overwritten.
It is very possible in Linux to have multiple versions of the same program. In fact I have multiple versions of GCC and even different versions of java (SUN, IBM, Blackdown).
Windows = The price tag will determine the amount of features you will receive with your choice
SUSE = One price for all features as cheap or cheaper than the cheapest version of windows + many opensource applications
For many users, Linux isn’t there yet. For many desktop activities like general office work, surfing the net and email, Linux is there.
Microsoft has also promised every time a new version of windows comes out, it’s more stable and more secure. I think the proof is in the pudding.
What do you thinks going to happen when a business examines an analysis of the books at the end of the year and realizes it spent x-number of dollars on the cost of the software, the cost of infections and the removal of them, the cost of down time, the cost of any lost data, the cost of lost gains, and the extra cost such as stolen data and the cost to their customers, and see they have a choice that doesn’t incur all these expenses? They may switch.
It’s not so much they product itself, but the company. Many people are sick of the MS treadmill and want to get off. I know of one personal example where a small local company saved over $3k in software licenses for switching over non-essential machines from Windows to Linux when they did a hardware upgrade. They had 3k’s worth of motivation to switch, imagine when a large company sees comparable savings.
I’m not saying Linux is great, all I’m saying is Microsoft better shape up.
XP is more stable than Windows 2000, XP SP2 is more stable than XP…
Where is the problem here?
Everybody has heard of Linux, and most have evaluated it to a degree that satisfies them. Most have chose something else over it.
Not only that, most have reevaluated it two or more times; since it’s always “. . . so much better than it was back then!”, then they discovered the same POS OS only with crap tacked on top of it so geeks can still claim it somehow “modern”. It’s still trailing behind in MANY key areas; but stallmanites don’t/won’t get it. You see, nobody can critisize the almighty Linux. Just wait until KDE/Gnome/X/Gizmogottahave version 0.x.12v is released “that’ll fix everything” . . . “It’s almost there!”
First impression are VERY HARD to change and thanks to the largest part of its vocal userbase: Everyone has had a first impression, before poor little Linux was “ready”.
So by time Linux is “ready” (big IF), no one will care or listen.
Everybody has heard of Linux, and most have evaluated it to a degree that satisfies them. Most have chose something else over it.
What kind of fantasy world are you living in. You are living in cloud cuckoo land. If you walked out of your door and asked passers by had they heard of Linux what percentage would say yes – pretty small. if you asked them had they evaluated it the percentage would be tiny.
The majority of people who evaluate it stay with it. When I first tried out Linux 5 years ago a good part of my usage at home on my dual boot setup was in Windows. Now I never boot up in Windows.
It’s still trailing behind in MANY key areas; but stallmanites don’t/won’t get it. You see, nobody can critisize the almighty Linux. Just wait until KDE/Gnome/X/Gizmogottahave version 0.x.12v is released “that’ll fix everything” . . . “It’s almost there!”
The point about Novell SLED is that they are releasing a Linux version that is not almost there but a version that is already ahead of Vista before Vista is relased.
I’d love to see MS make Visual Basic .NET (or even VB 6.0) specifically for linux to make ELF binaries and such with VB. (Sorry, I use Linux and love it, but I love programming in Visual Basic). Gambas, KBasic, and the like don’t cut it.
http://www.mono-project.com/VisualBasic.NET_support
I’m just starting to learn VB.net but I can get my work done on Linux with Mono Develop. It’s not exactly the same as there are some differences, maybe because Mono is more strict and VS allows some lousy coding. Mono Project claims that you can run VB.net apps compiled with Microsoft’s compiler with Mono tough.
keep on repeating things, and it will come true.
Microsoft doesn’t ship with dvd burning software either, or office software, or a significant number of drivers for that matter. We don’t have to emulate Microsoft. If we want to displace them, we have to produce a better product than them. Shipping with dvd support means that one more thing “Just Works.” One less headache. Besides, the legality of libdvdcss is dubious, I’d rather spend a very small amount and guaranty legality. It wouldn’t be very professional for Novell to direct their customers en mass to a piece of software that could be proven illegal a few years from now.
As I’m sure you are aware, having an RPM does NOT make it easy to install. That was the first thing I learned in Linux. We used urpmi with plf sources, great for taking the bite out of dep hell, but not good enough. I might get a new copy of Linux Format, and in reading the reviews, find an app I wanted to try. Half the time it was in plf, or could be installed easily though urpmi dep resolution, the other half required an hour or so of dep hell and a splitting headache. When in Windows it would have required nothing more that a few clicks. Even today, for software not provided by the distro, it’s often easier to just install from source, at least on RPM distros.
I had heard of the ease of apt, but attempts at installing debian ended in failure. At the time I had, like most people, nothing more than a moderate understanding of Windows. Half of the config options were completely unknown to me. Now I run gentoo, and have no dep problems at all, but we’re talking about new users. RPM is a distro tool to manage system software. It is nearly useless for day to day use.
Autopackage is at 1.0.9, btw, and works great on Suse. Which is what Novell cares about. All that remains now is for people to package their apps.
Autopackage is at 1.0.9, btw, and works great on Suse. Which is what Novell cares about. All that remains now is for people to package their apps.
And curiously, Novell is pushing their build system to make creating linux apps easier. You’d almost think someone within Novell sat down, drew up a list of why linux will never succeed, and has been whittling down the list.
By the way, regarding RPM’s, managing RPM with YaST isn’t really that bad, especially if you stick to the “provided” apps. I’ll grant that RPM has issues, but apparently you never had to deal with some of the more infamous windows .DLL files, where each vendor had their own set, and their versions frequently conflicted.
My current “available package” list (on SuSE 9.3, no less) has approximately 7,000 entries– a number of which are dependencies, -devel, and -debug entries, but still– easily several thousand applications in that list.
As for downloaded RPM’s, usually –test, and collecting all the necessary dependencies in one directory is sufficient to avoid most issues, until you come to “Gee, you have Xorg 6.8, and you need 7.0”. That’s when I bail, whether it’s Gentoo, Debian, or RPM based.
That when Linux is getting better, Windows and MAc also is getting better so Linux remain in the same spot..
Linux always gonna be like the balls of a Dog, behind!!
I’ve said for years.. it will be 2010 when Linux is really ready. It’s coming, getting closer and closer but it won’t be here this year or next. 2010 is the year of Linux. Be ready.
Some people seem to think that when Linux is getting better, Windows and Mac also is getting better so Linux remain in the same spot.. They’re wrong. Linux has some serious desktop issues to work out. It needs to get consistancy and better software distribution models and a framework for companies to provide their own commercial drivers easily etc. etc. but once you have something that is basically where Windows NT 4 / Office 2000 was Linux will be in serious play. I’m hoping that will be by 2008. Then, it will take a few years but people will start considering it less as a hobby os an innexpensive highly functional option.
People seem to make the mistake of assuming software’s progress will never slow, that software will never be good enough. As with all new technologies, computers will mature and settle down and the seemingly neverending flood of advancement of today will slow to a shadow of its former height. A new OS release taking 6 years will seem fast. Using 30 year old software will be commonplace. This is a world in which $300 office productivity software has a hard time competing with $0 software when they have the same features and have for 10 years.
Time is on open-source’s side.
Thank you for this dazzling John Titor inspired short story!!! A+!
in the future because the common custom of its users: never buy an application unless it is GPL (in short: I must pay to use software? Give me a break!). That makes developers reluctant to develop applications for Linux.
“never buy an application unless it is GPL (in short: I must pay to use software? Give me a break!). That makes developers reluctant to develop applications for Linux.”
Good thing Windows users dont ever think like this or else there would be horrible stuff like software piracy….
I don’t think Linux users are more reluctant to pay for software than people using windows. The difference is that the windows user tries to bring down the cost by using illegal copies, while Linux people use legal free software. I think this is particularly common for home users. E.g. I don’t think that many home users paying the full price of Photoshop, Illustrator, or even MS-Office. Linux users have more options of free software.
When Linux is deployed in a business setting there will be people willing to pay for software and support. If we look at the serverside, there are plenty of commersial software for Linux. There is no reason to believe that this would be any different for the desktop once it takes off.
As more Linux toolkits gets ported to other platforms you will see more free software on windows as well. OpenOffice, and Firefox is just a start. Software houses will no longer have a protected windows market. To compete they will need to sell their software to every user they can, including ones using Linux.
One of the big barriers to get Linux to the corporate desktop is the lack of cross platform calendaring tools. Companies will not accept multiple platform specific solutions for infrastructure as important as this as it makes planning of upgrades harder, there is also a risk that some bugfix upgrade to e.g. your Exchange server will create problems for clients from other venders. Once this obstacle is gone, I would think we will see a lot more corporat use of Linux and as a result there will be more people willing to pay for Linux software.
BTW, Novell could improve their chances on the on the business desktop by porting their PIM applicaton Evolution and their mail/caledar server Hula to windows. That way companies could switch gradually, with less worry for production loss due to lack of training on a new OS.
1) It’s bloatware…there are 900000000 distros thats only difference is name…this will only confuse potential consumer and make him run back to safer grounds…also, Linux distros “out of the box” functionality is rediculously limited…consumers want their desktops to “just work” after they install it…much like mac osx or windows. They dont want to spend hours surfing the web to find out what they must do to get their laptops hotkeys work
2) No big companies will support desktop linux because the money to be had on that market is limited because open-source activists refuse to pay anything for anything…Adope and such are companies and for company to be successful, they must be where the profit is at.
3) Opengl is dead…No sight of 100% native, directx compatible SDK anywhere to be seen.
4) Drivers, desktop user doesnt want to spend hours compiling a kernel just when they are switching from ATI card to nvidia card…the amount of drivers for hardware is limited and most of the drivers are pain in the arse to install.
Linux distros “out of the box” functionality is rediculously limited…
Hmmm. List what you can do with a Linux distribution functionally out of the box, like word processing, digital photo editing, image editing, mind mapping etc. etc. and then compare that to the list of software you’ll have to spend hours acquiring and installing to get the same level of functionality.
No system anywhere provides the same level of out-of-the-box functionality as an average Linux distribution, and nothing can do it cheaper.
consumers want their desktops to “just work” after they install it…much like mac osx or windows.
If you’d ever watched most users you would know that this is so untrue it isn’t believable. In no way do they just work. Beyond the base functionality of the OS, which is nothing, they have trouble with installing software, getting hardware to work – you name it. I think most people would love the huge functionality of open source applications pre-installed, or to have the drivers already in their system for a device, so that they don’t need to do any of that rubbish.
Adope and such are companies and for company to be successful, they must be where the profit is at.
Well, unless companies like Adobe do something they may find their market rather limited by open source applications once they get to the Linux desktop.
Opengl is dead…No sight of 100% native, directx compatible SDK anywhere to be seen.
Right…… So an accepted and well used standard is dead and people should be implementing the behemoth that is DirectX, which is actually an impossible amount of work to do? *Cough* bullshit.
Right…… So an accepted and well used standard is dead and people should be implementing the behemoth that is DirectX, which is actually an impossible amount of work to do? *Cough* bullshit.
Not arguing with your main point, but the Wine developers have made a lot of progress implementing DirectX on top of OpenGL. Many DirectX 8 and lower games work fine. DirectX 9 is a little more flakey…
Not arguing with your main point, but the Wine developers have made a lot of progress implementing DirectX on top of OpenGL.
They certainly have, and that’s OK. But the original poster was basically saying that OpenGL should be ditched for the standard that is DirectX.
Opengl is dead…No sight of 100% native, directx compatible SDK anywhere to be seen.
Jesus. Gamers are such morons…
1)
So, you’re anti competition ? The milk market is pretty confusing, so many vendors I wish there was only one because then I wouldn’t have all the hassle of making a choice, well guess I’d be more than happy to pay monopoly prices in return.
Apart from that, I recently had the joy of installing Windows XP. Wow, that was fun and fast. First it took ages to get all those security fixes and whatnot installed (remember the joke about what are the first n programms any user installs under windows) then came spyware blaster, spybot and ativir. Then, just to be on the usable side of things Firefox and then well then it was time to get the system to a state were you could actually use it (dvd stuff et all)
2)
Economists usually assume that people act rational. If a product offers enough benefit, people will pay for it. Unless you have strong evidence against this I’m willing to belive the same. Also I don’t understand your reasoning because we’re talking about regular users, right. Are you trying to say that if a user switches from windows to Linux he is suddenly going to become an “open source activist” ?
3)
Yeah, nc.
4)
Savy enough to switch the GPU but not to install the drivers, I doubt it. Apart from that, man modprobe.
You seem to know quite a lot about what users want, can you give some references ? Because, one might call me naive here, I think what users want is choices so they can see for themselves what works best. I have no evidence for that however apart from a sample size wich is certainly not large enough.
p.s. (in regard to another post): I don’t think Java qualifies as cross platform if you list the features of the latest version. Ask any PPC/Linux user and you’ll see that they won’t get to use those features.
Edited 2006-04-11 16:56
> p.s. (in regard to another post): I don’t think Java qualifies as
> cross platform if you list the features of the latest version. Ask any
> PPC/Linux user and you’ll see that they won’t get to use those
> features
True. There are some platforms it simply won’t run on yet. Hopefully, the Harmony project will change that. It does run on the majority of platforms that people are using though.
I’m not sure if Java is the real answer or not. What I am sure of though, is that going back to C++ as the first choice for new applications is not the answer.
Drivers, desktop user doesnt want to spend hours compiling a kernel just when they are switching from ATI card to nvidia card…the amount of drivers for hardware is limited and most of the drivers are pain in the arse to install.
This whole statement is BS. You don’t need to recompile your kernel when you switch your video card and even in the rare case that you did have to do that it doesn’t take hours. It only takes me 20 minutes and I have a 700Mhz processor. As for the ease of driver installation, Linux has the upper hand. Most of the drivers are built in, unlike Windows. Most of the drivers will be autoloaded when needed, unlike Windows. Most drivers do not require you to install them, in fact most of the time you don’t have to do anything.
This whole statement is BS. You don’t need to recompile your kernel when you switch your video card and even in the rare case that you did have to do that it doesn’t take hours. It only takes me 20 minutes and I have a 700Mhz processor. As for the ease of driver installation, Linux has the upper hand. Most of the drivers are built in, unlike Windows. Most of the drivers will be autoloaded when needed, unlike Windows. Most drivers do not require you to install them, in fact most of the time you don’t have to do anything.
You’re incorrect about Windows; in Windows, the drivers are auto-loaded; Windows reboots because the average coder at Widget incorporated can’t be bothered reading the relevant documentation on the Microsoft Developer Network knowledgebase.
Also, Microsoft has ALOT of drivers available, on Windows XP Professional (Service Pack 2), there are 78MB of drivers included with Windows; the only time I see problems are either when one uses a a bleeding edge, new piece of hardware OR when its an uncommon vendor.
With that being said, atleast when Windows installer does start, one can actually load extra drivers rather than with Linux, where its a matter of ‘if it doesn’t boot, and you need to load a module before hand, well, tough shit!’.
With Linux, the only gotcha with video card drivers; install the drivers THEN shut down, install the video card, then reboot.
Then again, one would assume that if one were to install a video card, one would also have enough technical skills to follow basic instructions in regards to Windows.
Most end users I see, don’t upgrade their video cards themselves, they get it done either at the shop or by a friend; but like I said, if one were doing it themselves, it would be right to assume they also have a reasonable level of technical knowledge.
You’re incorrect about Windows; in Windows, the drivers are auto-loaded; Windows reboots because the average coder at Widget incorporated can’t be bothered reading the relevant documentation on the Microsoft Developer Network knowledgebase.
They cannot be autoloaded when they don’t exist on the system. That was what I was getting at.
Also, Microsoft has ALOT of drivers available, on Windows XP Professional (Service Pack 2), there are 78MB of drivers included with Windows; the only time I see problems are either when one uses a a bleeding edge, new piece of hardware OR when its an uncommon vendor.
…or you are installing onto a very common sata hdd. In fact unless the computer is 4 or 5 years old you will most likely have to install at least three drivers on a XP SP2 install. The typical culprits are the video driver, the sound driver, and the ethernet driver.
With that being said, atleast when Windows installer does start, one can actually load extra drivers rather than with Linux, where its a matter of ‘if it doesn’t boot, and you need to load a module before hand, well, tough shit!’.
That would suck. Fortunately I haven’t ever had that happen. In fact it’s pretty rare these days.
Maybe what’s important about this interview isn’t what was said – nothing new there – that that it comes from Ron Hovsepian, who is an IT pro, rather than from Jack Messman, whose last job was with the railroads, perhaps ensuring that trains ran on time. If this indicates a change at the top at Novell, with a more IT-savvy guy moving to take full charge, then well and good. Not before time, either. Just guessng.
also, Linux distros “out of the box” functionality is rediculously limited
And “out of the box” installation of any other operating system is usually a lot more limited, because a Linux distributions is not just an operating system, but operating system plus end user software.
because open-source activists refuse to pay anything for anything
That is a grave misunderstanding of the market. The users refuse to pay for software that is not as good and often worse than a free (as in gratis) alternative.
Software which is better than its competition will always gather buyers, but if a companies think consumers will buy it just because of the product name, they will be disappointed.
If Nero for Linux had the same functionality as Nero for Windows, it would be bought more often, but it isn’t and thus most consumers don’t get any added value by buying it.
Opengl is dead
Hehe, that’s a good one
Hint: once you reach the high end 3D sector, OpenGL is the option you can rely on, everything else is marketed as an additional feature.
If Nero for Linux had the same functionality as Nero for Windows, it would be bought more often, but it isn’t and thus most consumers don’t get any added value by buying it.
I doubt it. There are too many free alternatives for Linux that allready are better than Nero for windows.
The problem for Nero is that they got too late to the market. If they had been present on the Linux market earlier there would have been less incentive to develop good free programs like K3b.
This question may have been discussed on forums. i dont know right answer..
Is it Illegal to develop closed source applications for a open source GPL linux kernel??
Because some or other day hardware OEMs will face it for developing linux drivers and also Novell, RH in for their desktop?
I don’t care when Linux will be ready for the Desktop. It is ready for me.
install the drivers THEN shut down, install the video card, then reboot.
While I don’t think this is a big problem at all, it would be nice if the X server tries vesa mode in case it’s not able to load driver listed in xorg.conf, instead of immediately falling back to the terminal.