Nat Friedman discusses Hula and other linux issues on the latest episode of The Linux Link Tech Show: episode 71.
Nat Friedman interview on The Linux Link Tech Show
About The Author
Eugenia Loli
Ex-programmer, ex-editor in chief at OSNews.com, now a visual artist/filmmaker.
Follow me on Twitter @EugeniaLoli
Evolution for OSX (non-native)
Talks about XGL @ about ~24mins
The state of H/W support in Linux
More attention to Gnome in SUSE, (G2.10 in SUSE9.3!)
Been a fan of the show for alittle bit now, it’s pretty good. One of the best features is the interaction with with the people in the chat channel irc: http://www.thelinuxlink.net channel #techshow each wednesday 8:30pm. The channel seems to fill up each time that week and its pretty fun. I sort of feel like an advertisement but I had to give these guys props
Summary:
<li>Apparently he’s not involved in Novell’s Linux desktop development any more. That was his function until recently. He’s now basically working on Groupwise (what he calls Novell Collaboration Teams), but basically says he’s working on an an array of software and Hula came out of that.
<li>He’s stirring around, trying to find something that Google has got.
<li>The plugin structure for Evolution is the coolest thing ever. Just a pity it isn’t actually the most popular, or even the second most popular, e-mail and groupworking client around.
<li>They’re porting Evolution, D-BUS and every library it needs to Windows. The question why wasn’t asked.
<li>The 3D X server is cool, and he waxes lyrical that you don’t get overlays with Windows when you move them around. This is probably more important for Gnome and GTK than anything else. However, it absolutely needs 3D hardware, and it needs an nVidia because ATIs tend to crash quite a bit.
<li>Subject of Suse as a KDE oriented distro was brought up. Nat butts in: “Whoa, we’ve changed that”. Have you? No one else has noticed. “You should check out Suse 9.3” says Nat, just as he said about 9.0, 9.1 or 9.2.
<li>The subject of education was brought up, and apparently there are no educational apps. They obviously didn’t look at something like the KDE Edu module.
<li>Apparently, a broad array of technology is going to find it’s way into Suse 9.3. I’ll wait for the interview with someone from Suse for that one.
David, you’re approaching epic proportions with your anti-Ximian trolling. I don’t know what your problem is, but you appear to have an immense hatred of anything Ximian. I don’t have the inclination to rebut your moronic ramblings, but seriously: get a life.
first off, microsoft isnt developing .net to sell copies of vs, or their implementation of it. sun isnt selling java, or netbeans/forte/sunOne. but for some reason novell isnt being smart because its doing the exact same thing as everyone else?
as for groupware, upper management adores it. jamie hates it, but jamie hates most things that have to do with corporate infrastructure. just because jamie doesnt like groupware has no bearing on the fact that it is an essential componant (or at least perceived to be, but thats besides the point) of any serious business. just because you dont find it useful doesnt mean it isnt.
ximian was a company, they were not a non-profit orginization. companies make products then go out and try to sell them. this is what ximian did. i would like to know how something like evolution is not vital for a “successful business model” unless you ignore where you are going to get income from. if you are selling to people who wont even consider a product that cant connect to their exchange server, then ignoring them and working on cool things (the kind of stuff hackers like jamie like) what you would end up with is a company like eazel, not a company that gets bought by a player like novell because of the technology it has.
everyone hypes stuff, the bigger it is the more it gets hyped. ximian is not vaporware. and ximian stuff (like evolution) is a big reason that gnome is the desktop that pretty much every company chooses to deploy, you may consider that rubbish, but i dont.
Get your timeline straight. Suse 9.0 was released before Novell acquired Suse.
Hula isn’t new code, as it’s based on the proprietary Netmail product. Would you have preferred it if Novell had just thrown the code away?
> Ximian and the people from it had an absolutely
> fantastic opportunity to do it, but they got blinded
> by other unimportant things and didn’t think through
> how they would fund themselves.
Are you claiming that all Ximian software is rubbish because the company was acquired?
Yes, take Mono. Is the project worthless because you’re foreseeing future funding issues? Perhaps you should enlighten us as to the future fiscal realities of other projects. Will funding be a problem for the kernel? For X? For gcc? Or are only Ximian projects affected?
The only reason Trolltech’s licensing works is because they’re the only ones doing it. Imagine what would happen to the viability of commercial software on Linux if developers had to pay for each level of the software stack (kernel, gcc, glibc, X, toolkit, desktop environment).
Get your timeline straight. Suse 9.0 was released before Novell acquired Suse.
Exactly. It was around the time that Suse was acquired, but before January 2004 when it was completed. Why were the Ximian people giving the impression that their desktop would be tightly stitched (in their words, not mine) with Suse Linux 9.0?
Hula isn’t new code, as it’s based on the proprietary Netmail product. Would you have preferred it if Novell had just thrown the code away?
They could have made it work properly before releasing it, or made it integrate better with existing open source groupware systems and standards. There are a few. It will take years to get it to work in the way they describe, and it currently doesn’t warrant the hype.
Hopefully now that Jamie has given Nat a reality check it will head in the right direction.
Are you claiming that all Ximian software is rubbish because the company was acquired?
No. It’s their positioning and thinking behind it that was always the problem.
Will funding be a problem for the kernel? For X? For gcc? Or are only Ximian projects affected?
The Linux kernel and gcc are low-level projects that are worth the investment – Linux servers that require them are big business, and they’re not overly large and complex pieces of software. They actually pay for themselves. X has been around for many years and isn’t as complex as a desktop environment or a whole toolkit for a desktop environment.
The closer you get to the desktop the more you have to think about the complexity of what’s involved, the development tools needed, the investment that is obviously required to make it good enough and how to go about doing it. That’s another unseen hurdle for desktop Linux in general, and unless someone somewhere understands that then everyone is blowing in the wind.
Imagine what would happen to the viability of commercial software on Linux if developers had to pay for each level of the software stack (kernel, gcc, glibc, X, toolkit, desktop environment).
If they were too complex and required too much investment for them to be free in every sense then I’m afraid you would find yourself paying for them. When companies and people pay for a Linux distribution or a Linux server they are actually paying for those projects and the contributions to them, however small.
and ximian stuff (like evolution) is a big reason that gnome is the desktop that pretty much every company chooses to deploy, you may consider that rubbish, but i dont.
I’m afraid I see no evidence of that and never have done. If it were the case, then Ximian didn’t do a terribly good job of selling support when they were around.
as for groupware, upper management adores it. jamie hates it, but jamie hates most things that have to do with corporate infrastructure. just because jamie doesnt like groupware has no bearing on the fact that it is an essential componant (or at least perceived to be, but thats besides the point) of any serious business. just because you dont find it useful doesnt mean it isnt.
Read the article. Read what Nat wrote on his blog and/or hear what he had to say on this talkshow (basically, more or less the same except for some details). Seriously!
1a) Jamie’s argument wasn’t that he hates groupware. He thinks groupware software is aimed for the wrong aims (not enough for calender) b) was rather meant to impress managers b) than be practical to its users for who it should count. Those were his main 2 arguments. He also makes various interesting references to his former employer.
2a) Of interest, is that Nat got convinced by Jamie AFAICT (based on both the interview and his post on his blog) b) also, if i read correctly, JWZ actually made the argument he made a while _earlier_ (to Nat / Novell) but published the notes only now.
i dunno, i still disagree. maybe its just me, but meetings and whatnot tend to be more beneficial to management then to us developers. we only use outlook as little as we can get away with. any benefit you get from design meetings, or update meetings can usually be acomplished far more efficiently over a smoke and coffee 😉 the target users are management, and stuff like outlook, lotus, or evolution is pure heaven for them.
The comment/question was not that there are no educational applications in Linux, but compared to the volumes of educational applications available for Windos and Mac OS and Novell having a strong presence in K-12 education, whether this might influence software developers to port some of these educational applications over to Linux.
I specifically stated that the educational offerrings currently available in Linux are outstanding.