Filed in archive
General
by Scott Wilson on March 1, 2008
The issue, briefly, is one faced by many companies: critical, but ancient, software running on failing and irreplaceable antique hardware. In the case of the Seattle School District, the software is their carefully devised student placement system, and the hardware is a lumbering old VAX (obligatory nostalgic moment: my first college BASIC course required programming be done on the school VAX, although my home PC at that time probably had more horsepower).
Nick suggests that this is likely a common problem for school districts, and wonders why the open source movement can't address public problems of this sort, with its vaunted volunteer community and can-do attitude. Why not make the requirements public and let the community take a shot at them?
Although my own thoughts also sprang immediately to open source, I confess they moved in a different direction, and went roughly from "There has to be a free, open-source solution to this problem" to "Yeah, I'm sure there is, and the district IT department is stuck in the same rut as every other large IT department and hasn't bothered to look for it. Millions of dollars to fix, my foot." And I was so certain of this that I didn't even bother to go look it up, until I read Nick's post, and of course there is.
Which got me wondering, is it the fact that so many people think in the same terms as Nick that they avoid adopting open-source solutions in the corporate world? Because while the advantages he suggests are real, there are inherent drawbacks to the development method that any right-thinking CIO would be considering at each point; from experience, I can tell you that volunteer development efforts don't necessarily result in anything positive... go cruise Sourceforge and figure out the ratio of projects you've heard about to those you haven't, or even those that work to those that don't. While Nick's suggestion sounds good, and may even be good in the fullness of time, I know a lot of IT executives that would run quickly the other way at such a suggestion, and they probably would not be wrong to do so.
The thing is, there is definitely a place for open source solutions, and they can be quite excellent and cost-effective. So is the community shooting itself in the foot by making things more grandiose and complex than need be?
I feel a bit of a cad, though, as Nick has volunteered his time to architecting an open source solution for class placement. I, therefore, volunteer to donate one of the old 486s laying around in my basement to run the emulator software on. No, no, don't thank me for the hundred-fold speed increase or the two-day implementation timeline. Just doing my civic duty.
[Disclaimer: For the anally retentive readers, yes, I know there may be technical issues with the specific emulator I linked to, it's just an example; there are a half-dozen available. And I know a full-fledged, native application would be a better solution for school districts, but I also know the budget reality. And don't even talk to me about the details of figuring out what is and isn't a "working" project on SourceForge, the point and plain fact is that there are more abandoned or undeveloped open source projects than successes]
Permalink: An Open Source to cure all ills
Trackback: http://publish.creative-weblogging.com/publish/mt-tb.pl/115480
Addthis
Ask
Blinklist
del.icio.us
Digg
Fark
Facebook
Google
Lycos
Ma.gnolia
Mr Wong
Netscape
Netvousz
Newsvine
Reddit
StumbleUpon
Slashdot
Tailrank
Technorati
Wink
Yahoo
Vote for An Open Source to cure all ills:
|
Rating: 9.50 out of 2 vote(s) cast.
|
Response from:
Kiat Huang
(03/01/08 5:35pm)
Response from:
Scott Wilson
(03/03/08 10:53am)
I think you should avoid criticising anyone else's logic when the best rebuttal you can come up with consists of "surely" and "I suspect." :)
As you say, the numbers are not generally available for proprietary projects. But in my experience they are abandoned less frequently; some organization is making an investment, and (if not as often as they should) they have run the numbers and know they have a business need for the solution, and have a good idea of the resources they can commit to it before they even start.
The failure number is still extraordinarily high--but lower than in the Open Source community.
The "unrestrictiveness" you speak of is exactly the point; it sounds good if you are the developer, but the user WANTS some accountability and predictability... a point frequently missed by open-source advocates.
If I might offer you a tip, presuming you're an open-source fan: don't try to defend or draw contrast to project abandonment rates between open and closed source projects. Instead, point out that the nature of open source allows more projects to be started with fewer resources, and that the beauty of the system is that these can be consolidated or discarded with little ultimate waste and potentially to some future advantage in lessons learned. As it is, you're making your argument according to ground rules which are already stacked in favor of more traditional proprietary development approaches, and even if you had all the data you need, you'd find yourself losing it. It's acceptable for open-source to be "better" than proprietary software in unconventional ways; you don't need to justify it by traditional metrics.
As you say, the numbers are not generally available for proprietary projects. But in my experience they are abandoned less frequently; some organization is making an investment, and (if not as often as they should) they have run the numbers and know they have a business need for the solution, and have a good idea of the resources they can commit to it before they even start.
The failure number is still extraordinarily high--but lower than in the Open Source community.
The "unrestrictiveness" you speak of is exactly the point; it sounds good if you are the developer, but the user WANTS some accountability and predictability... a point frequently missed by open-source advocates.
If I might offer you a tip, presuming you're an open-source fan: don't try to defend or draw contrast to project abandonment rates between open and closed source projects. Instead, point out that the nature of open source allows more projects to be started with fewer resources, and that the beauty of the system is that these can be consolidated or discarded with little ultimate waste and potentially to some future advantage in lessons learned. As it is, you're making your argument according to ground rules which are already stacked in favor of more traditional proprietary development approaches, and even if you had all the data you need, you'd find yourself losing it. It's acceptable for open-source to be "better" than proprietary software in unconventional ways; you don't need to justify it by traditional metrics.
Response from:
Matt Burkhardt
(03/04/08 9:23am)
There's an incredible wealth of open source software for the K-12 market. I'm constantly amazed and surprised by the breadth of applications available - from Student Information Systems to Integrated Library Management Systems - that are available. Indiana schools decided to move to open source software and it has and continues to be a resounding success. Check out http://www.imparisystems.com/index.php?option=com_content&task=view&id
=32
for some of the free and open source SIS's available.
=32
for some of the free and open source SIS's available.
Response from:
Stanley F. Quayle
(03/08/08 3:01pm)
I've spoken with the IT people to see if they'd be interested in migrating to an emulated VAX, which my company sells (http://www.stanq.com).
The response was system availability wasn't an issue.
There's an open-source version, called SIMH. But moving to that will violate their license agreements -- HP now owns VMS.
Instead, by migrating to CHARON-VAX (the product I sell), they can get legal transfer licenses from HP for $2k.
Either way, they'll move to a 3U size PC in a rack, instead of the 3 or 4 racks they're using now...
The response was system availability wasn't an issue.
There's an open-source version, called SIMH. But moving to that will violate their license agreements -- HP now owns VMS.
Instead, by migrating to CHARON-VAX (the product I sell), they can get legal transfer licenses from HP for $2k.
Either way, they'll move to a 3U size PC in a rack, instead of the 3 or 4 racks they're using now...
Subscribe
Marketplace
-
Online MBA Degrees - earn your mba degree online with one of hundreds of programs available at elearners.com
Use the search to look for other interesting posts
| RSS |  See all blog subscribe options |
 What is RSS? | |
| Yahoo! |
|
| Addthis |
|
| Bloglines |
|
| Newsletter | |
| Follow us on Twitter! |
It is interesting the writer (Scott Wilson) uses flawed logic when looking at the distribution of the health of Sourceforge's open source projects - there are surely just as high (and I suspect a lot higher) a proportion of dead propreitary projects on developers floors and in office bins. You just don't hear much of them. The "proprietary community" (but of a misnomer: since they can't really engage in candid, unfettered code discussions) in general cannot learn collectively from their own member's mistakes...and if a community cannot learn efficiently from it's own mistakes, then it fights a losing battle in the long run, with a viable alternative that really can.