Why SATV doesn’t use Linux, Part 2

Continuing my rant on Linux, with a few more of Wolfe’s points and some of my own:

Disclaimer:  Salem Access Television has no institutional opinion on Linux, and neither does its staff, board or membership.

3. You can’t make money on the operating system

Wolfe quotes from an article from builder.au by Con Zymaris, How To Make Money from Open Source,

Marketing your offering
What does it take for your product to get noticed? Obviously, by using open source attributes when marketing your product: Market the fact that you give your customers the complete source code to the system; market the fact that the code does not have a use-by date or sunset clause. If you and your business collectively fall under a bus, your customers can continue to use and have third parties provide ongoing support. Leverage the fact that local business and government consumers are risk averse, and that you, unlike a group of coders in Iceland or Brazil who produced the original codebase, can indemnify your customers using your professional and product liability insurance; market the fact that you are local or regional and can provide same time zone business support.

By commercial support, I mean commercial support. Charge the customers $200 per hour for it, but make sure you deliver the goods. You should also play the perpetual code escrow card. Many potential customers of vertical business applications need to be guaranteed that they will not be left stranded when deploying a new line of business system.

OK, say we do get a Linux-based solution in our field, and we get code escrow for our Linux-based app and our consultant happily forks (changes) the code.  And he is run over by a bus?

Then we have to find some other consultant in our field.  Neither I nor SATV have the resources to maintain the code ourselves;  it is not our business.

How is this really different from proprietary code?  At $200 per hour? 

SATV pays for its software;  we don’t get it from Bittorrent and we do expect to incur licensing and support costs.  But we are a small nonprofit with an equally small budget to match.  When a product costs us that much to have supported, I must be discerning and look carefully at the TCO just as much for open-source products as for proprietary.

The meme around open source is that support is "special" and doesn’t need to be considered just because it is open source.  Open source software is subject to the same "laws" and market forces as proprietary software and no amount of ideology can change that in my mind.

There are other points that Wolfe and his commenters touched upon that I want to address:

"Well, you can subsitute <y> for <x> in Linux!"

The Linux community often hypes substitutes for proprietary software.  "You can run Open Office instead of Office!"  "The GIMP instead of Photoshop!"  "Run WINE to emulate your Windows apps!"

In my other life, I’ve designed graphics for broadcast at SATV for 12 years.  I’ve used The GIMP for 9 of those years.  I loved it.

Today, I’m all but ready to dump it and get Photoshop or at least Photoshop Elements.  There have never been a large number of GIMP users, and while the GIMP team has served me well for 9 years, I’m missing the potential of Photoshop.  All those Photoshop techniques that I can’t adapt to the GIMP have caught up with me.  I’ve been using the same techniques as an artist that I used 10 years ago and I hate it.  Tools are very important to creatives;  for example, Final Cut Pro will not make you a brilliant videographer, but if you’re merely a good videographer, iMovie will only get in your way.  The GIMP is in my way. 

I don’t look forward to having to budget even for Photoshop Elements but I don’t feel I have a choice.  Elsewhere at SATV, I tried to make GIMP our standard, but at the time it integrated poorly with Windows and I couldn’t get the staff to be comfortable with it.  We bought Paint Shop Pro.  It cost us, but at least our staff could get work done without tinkering.

Linux application Y is not like Windows application X.  At best, it can be a very good usable copy.  At worst, it can be an unusable knockoff.

And WINE?  Sure.  I’m going to add another unknown abstraction layer to our software.  I might use WINE if I loved my distro and I loved even more that shareware app that only existed under Windows, but I don’t love Linux enough to inflict that kind of pain on us.

Computing for People with Disabilities

Through personal experience, I’m sensitive to the needs of people with disabilities.  SATV has been host to our city’s Commission on Disabilities;  I have been their liason to SATV for 8 years and just recently became a commissioner myself.  We have staff, interns, volunteers and members using our computers and they must be usable by a wide range of people.

I looked at Ubuntu recently;  Ubuntu is probably the most well-realized desktop Linux distro to date.  It has accessiblity features for people with disabilities, including a magnifier, an on-screen keyboard and a screen reader. 

I have low vision, so I often use a screen magnifier.  I installed Ubuntu 6.06 LTS on Virtual PC and tried it out. 

It was terrible!  There’s a six-letter term I wish I could use if it were only work-safe!

First, some background on the Windows magnifier:  When it’s activated, it sits on the top of the desktop like this: 

You can configure the magnifier settings.  The only bad part is that it disables ClearType and interferes with the Aero desktop.

This is Ubuntu’s Gnopernicus:

The magnifier is on the top center.  You can’t adjust it and it covers any icons you may have on the upper part of the screen.  It looks terrible.  I’m mystified by this, as screen magnifiers are not a new technology and I know of several Windows freeware magnifiers that are perfectly useable.  I can only wonder if its shortcomings are due to the (ancient) X Windows subsystem.

Windows has a very useful feature that helps me avoid using a magnifier:  Font sizes can be scaled to any arbitrary figure.  This has existed since Windows 95 and for 12 years I have run Windows with its fonts scaled at 130%–virtually unchanged across Windows 95, XP and Vista! 

Ubuntu doesn’t have this.  I can, of course, change each of the fonts, but based on my experience with SuSE a few years back, it would only make things worse as I upset the proportions.   That’s a deal-killer for me.

At least Windows has JAWS and ZoomText.  No Linux versions, though.

Conclusion:  Just too many risks upon risks

Going to Linux on the server, let alone the desktop, would not simplify my job.  At a minimum I would still have to support Windows and Macs.  I’m not going to add a third major platform just to satisfy ideological concerns.

I’ll gladly consider using Linux for certain things, like firewalls (m0n0wall) or monitoring (Groundwork Monitor), that Linux is best suited for anyway.

But not as an ideology or a religion.



7 Reasons Why Linux Won’t Succeed On The Desktop — InformationWeek

3 Comments on “Why SATV doesn’t use Linux, Part 2”

  1. Unknown says:

    I think that your views on Ubuntu may be quite a bit off.You yourself said that you were using Ubuntu 6.06, which is almost a year and a half old. Unlike Windows, which took 6 years to get its latest update, a new version of Ubuntu is out every 6 months.Plus, Linux is evolving pretty quickly. Basing your arguments on old Linux software is just as ridiculous as bashing Microsoft in 2007 for Windows 98.Try Ubuntu Gutsy (7.10). It comes out tomorrow. Re-review it then.

  2. David says:

    I am going to grab 7.10 but it may not change my mind.  The sticking point is Gnopernicus, the screen magnifier.  According to its website, it\’s still at 1.0.  6.06 LTS cannot be that old!  The whole point to 6.06 is that it would be supported long-term to \’09.  In a serious desktop deployment, it would be the distro of choice.  Businesses don\’t much like having to do forced upgrades more than once every few years.
    It\’s slow to you or I conditioned to get the latest torrent, but not to the business I work in.  If XP didn\’t work on my machine (irrelevant in my case but just suppose), would you suggest Vista?  *cough*
    Either I have to believe that Canonical forked Gnopernicus and kept the codebase to themselves (which I find hard to believe) or there really hasn\’t been any development to 1.0.   And if there are bugfixes, is Canonical really going to say they can\’t put them in 6.06?  I don\’t know what development needs to be done to Gnopernicus to make it fit my needs, but it looked for all the world like a version 0.5 alpha.
    The sad, and disappointing part, for I like what Canonical is doing, is that I could try Feisty and it still may not work for me no matter how many times I hit the torrents for Gutsy.  That\’s just life.

  3. Rick says:

    If you don\’t count the stream of patches to Windows XP, then yes, it\’s true there were 6 years between major OS revs.  Incidentally, the beta for XP SP 3 has just started up.
    As for the age, well, a lot of businesses were just moving to XP a couple of years ago.  For some reason, businesses like stable OSes and don\’t want to replace everyone\’s OS every 6 months.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s