40 comments

[ 3.6 ms ] story [ 74.4 ms ] thread
Just what we needed, a timely contribution to help bring Linux to the desktop. /sarcasm
Maybe soon we'll have open source Motif, then I'll finally be able to not give a shit.

Funny timing though, I'm pretty sure I saw a Motif app or two on one of the screens at NASA during the Curiosity landing.

What is wrong with Open Motif: http://www.opengroup.org/openmotif/ ?
If you'd actually read the page you link to you'd see the part where it says "Read the faq for clarifications on the license... such as why Open Motif does not meet the Open Source definition."
Finally! Now I can hope for fat window borders on my phone!
Excellent, was always my favorite GUI as was compact, fast, and could get on with things with the only real distraction being xeye's that you could get bored of very soon.
You what? I worked on a competitor to CDE. We existed first and were actually compact and fast. When CDE was announced it put a serious dent in our business, but then was vapourware for quite a while so our sales returned. And once available was bloated, ugly and slow.

We also had a side business in the graphical toolkit CDE used (Motif). Basically we took the Motif from OSF, fixed bugs and made a whole host of similar improvements and then sold and supported the result.

http://en.wikipedia.org/wiki/IXI_Limited

Sorry to hear CDE hits a nerve with you. Vaugly recall your product from my days at DHL and when we looked at UNIX on the desktop with SCO, but SCO etc priced themselfs out of the market on that one on comparision to windows.

So what happened to your GUI today?

As for Motif, well OSF is the Open Source Foundation right and if so selling open source, can see how that business model worked out.

IIRC your GUI was mostly desktop/PC based and in that CDE wasn't a competitor so i'm somewhat confused, but I belive you.

Open Software Foundation (or "Oppose Sun forever"), they were never about "open source" but was a corporate group originally formed to fight AT&T/Sun and later Microsoft/WinNT. Despite millions of dollars investment in R&D and product development, they didn't produce anything of value. They also failed miserably in competing against Windows-NT. Except BSD, none of "traditional UNIX" had much to do with open-source.

http://en.wikipedia.org/wiki/Unix_wars

http://en.wikipedia.org/wiki/Open_Software_Foundation

http://en.wikipedia.org/wiki/USL_v._BSDi

While we were a division of SCO, SCO weren't that relevant to us. They bought IXI to ensure stability in supply of their desktop. SCO distinguished itself from other Unix vendors by having a friendly desktop including for administrators and their tools.

IIRC our business was something like 3% of sales on Intel chips with the rest being RISC, although SCO itself isn't counted in that since we didn't sell on the platform (corporate took the software and customized it to their needs).

SCO definitely considered Windows (NT) to be the competitor. However SCO was considerably cheaper than the RISC workstations and operating systems. The changing usage patterns are what led us into the browser space, into making Motif look like Windows so when displayed on Windows it wasn't jarring, and then our foray into letting you access any app via your browser which became Tarantella.

OSF (you didn't get the expansion quite right) was a pawn in the Unix wars: http://en.wikipedia.org/wiki/Unix_wars GUI issues in the Unix variants were the battleground in the early nineties, and finally petered out in the late nineties when Windows "won".

Our (IXI, not the SCO corporate parent) major product of the time was X.desktop which only ran on Unix systems using the X window system. ie it was Unix based apps to make Unix easy to use.

We then moved into "client access" which meant terminal emulators, X servers for Windows and file access (VisionFS) which were Windows based programs to make it easy to use Unix systems (except VisionFS).

Finally we did Tarantella which made it easy to use apps from just a web browser, and it was irrelevant if the apps were Unix or Windows.

Tarantella was eventually bought by Sun which was eventually bought by Oracle, but the product is still for sale. Some details at http://en.wikipedia.org/wiki/Sun_Secure_Global_Desktop (Disclosure: I was the architect for the non-gui protocol parts of Tarantella)

VisionFS is what I am most proud of - details at http://www.rogerbinns.com/visionfs.html

I like that it's on Sourceforge, as though they're inviting people to ignore it. That said, I'll probably try building it over the weekend, just for nostalgia's sake.
what's amazing is that something like CDE's default look seems very "clean" and utilitarian in a sense, after so much exposure to modern desktop environments... it's like the computer screaming at you: "it's about getting things done, Sticky, not about aesthetic masturbation!"...
I read this as:

"CDE open sourced..." <audience laughter, comedic pause>

"...on SourceForge" <riotous laughter, fade to black>

Pardon my ignorance, but what's wrong with SourceForge? Is it simply not as popular as other open-source project sites?
Obnoxious ads everywhere, download page with countdown timer, and as you mentioned the OSS community has largely abandoned it. Do you think this doesn't matter?
Sourceforge was great back in the day when it was the only (?) option for public code hosting besides running your own site, but it's not as "cool" now... it feels like most projects have left, only abandoned projects remain. Putting a "new" project there seems strange. I don't know if that's due to functionality or not though, maybe it's just that the UI feels dated. Even with free offerings, people pay attention to the usability and UI, and although I've never run a project using sourceforge, I don't think I'd ever choose (or recommend) it over github or bitbucket. I'd love to hear reasons why I'm wrong on that opinion?
I had a conversation with a senior engineer on SourceForge where he pointed out that most people forget that developers are actually a minority of their business: most of their users are nontechnical people that come to them to download binaries of things like zip utilities.
It may be true that download users are their business, but developers certainly are the drivers of the site. They produce the content.

Even then, the downloading experience is warped straight from the 90s.

There are valid reasons not to use sourceforge.

You know, each sf project can have a "homepage" link, and I once spent half an hour searching for where to change it. I kid you not.

I have searched for 10 minutes on how to reply to a bug report and attach a patch. I gave up in the end.

I'm still the owner of a dead project on sourceforge. I've tried several times to delete that project, but there was always something that prevented me from doing it (I don't recall, I think the project had a mailing list, and I didn't have permission to delete the mailing list, and I couldn't delete the project without first deleting the mailing list. Or something.).

sourceforge simply isn't usable. The makers know that, which is why they do a big redesign/relaunch every two years or so -- but each one makes it worse.

Do you know any alternative to SourceForge?

SourceForge allows to host phpBB, Wordpress, MediaWiki, my custom php code, with gigabytes of disk space.

It's not possible to install all this software on Github or Google Code project.

Perhaps that's why SourceForge is so bloated; it's trying to wear too many hats. GitHub and Google Code both give you a issue board, wiki, and site host; but it's all integrated and their ops teams don't have to worry about random 3rd-party software going haywire since they own it all.
LedgerSMB is on Sourceforge and probably will never move. Moving is a lot of work. Sourceforge has some quirks but it isn't really all that bad. Even for projects on GitHub, it's not bad to have a Sourceforge presence for hosting downloads and the like. It is true that in general we have been trying to move more and more of our services off Sourceforge, but this has been a slow process.

Sourceforge offers a lot of things that GitHub and Bitbucket don't offer. I haven't been able to find email list hosting integrated with project management anywhere else yet but maybe I haven't been looking hard enough. The closest thing you have is Google Code.

I wouldn't say there's anything wrong with it. It isn't as trendy as some others, but I think you want to look for concrete benefits, not worry about where the hottest spot is at the moment.

* Bog down slow.

* Unhelpful user interface with a bad/non-existing thought to workflow.

* Annoys the user.(You want to download that 30k file, please wait 5..4..3..2..1..).

Bad memories of writing CS assignments on overtaxed Sun workstations, with that crap WM. I fetishized those expensive purple boxes until I had to use one.
SGI Indys were my purple box fetish.
You mean Indigo, the Indy is blue :) (mine doesn't work anymore, alas, but the Octane is doing great).
Exactly the same experience here ... Can't say I'm a big fan of CDE, but it was obviously from the previous decade ...
You mean you didn't just use XDM to switch to a different environment like everyone else in the know? I just switched to FVWM, which was my favorite window manager at the time.
CDE... Oh the memories.

This is a good thing. Those few unfortunate souls who have to care will now have free access to the code in order to figure out the problems they are having.....

I remember when CDE was awesome. It was not during this decade, er, century...

I think it would be useful for us to all pause and think about what being on the right side of history looks like. CDE could have gone open source back in the 90s and been extremely relevant and useful. I'm glad it's happened because I think there's value in having an accurate and complete software archaelogical history...but I doubt anyone will be digging into this as a project to do current work on.

Slightly off-topic:

What if they open sourced Open Step? Is that even possible to do? Would that change open source Un*x desktops forever or would no-one care?

I doubt that it would have significant impact. There is already GNUstep which is GNU implementation/clone of OpenStep.
I wonder if this might actually be useful for people on very restricted (or legacy) hardware? How does it stack up against eg: LXDE?
Speak for yourselves, I'm excited! Unity and GNOME and all the other useless crap civilisation has invented over the last decade and a half? Get off my lawn!