Linux Today: Linux News On Internet Time.
Search Linux Today
Linux News Sections:  Developer -  High Performance -  Infrastructure -  IT Management -  Security -  Storage -
Linux Today Navigation
LT Home
Contribute
Contribute
Link to Us
Linux Jobs


Top White Papers

More on LinuxToday


Editor's Note: Teaching Old Dogs New Tricks

Oct 28, 2005, 23:30 (2 Talkback[s])
(Other stories by Brian Proffitt)

By Brian Proffitt
Managing Editor

Old habits, they say, die hard.

I am unsure who "they" are, though I know it's not the voices in my head, since they are not usually so wise. But regardless of their identity, they seem to be mostly right in this case.

For instance, I lost the habit of eating breakfast in the mornings in college, when I would have a choice of getting up and walking across the entire campus to the Union in the freezing cold or staying in my nice toasty bed an extra hour before Western Civ. Four years of such behavior beat the breakfast urge out of me.

So, too, do members of the IT community tend to cling to the old ideas that seemed to succeed for them or others in the past. It seems a sound bit of logic: if it worked for Company X, then it should work for my company, right?

Maybe. But the hardest thing to know is when to hang onto ideas that will truly work, and when to drop them in favor of a new innovation that will work better.

Take, for instance, Novell. Novell is a company that is not doing so hot right now, and I really hope that changes. On a technical level, I think they have a good products to offer in SUSE and Ximian, and a huge amount of experience (both good and bad) in the reseller channel that will only benefit the growth of Linux. But one thing they do that puzzles me (and quite a few other people I have spoken with) is hang on to the old per-user licensing model.

I am no economics major, but it seems to me that using the same type of licensing model as many of the other proprietary companies--even if your's is less expensive--is a tricky path to take, especially when your product is open source. Novell, to their credit, may not have much of a choice. They are trying hard to win back the value-added resellers they ticked off a few years back when they opted to bypass the resellers and go more direct market, and a per-user pricing structure would keep the VARs more happy. I think.

I am wondering, though, if other pricing structures, that would be at once unique amongst proprietary competition and beneficial to Novell are possible. It's open source, folks... maybe it's time to think out of the box.

Such a notion is certainly true regarding the migration to Linux in the German city of Munich. There, city IT officials are citing the problem of migrating 300 or so custom apps over to Linux. That's no easy hurdle to jump, I grant you, but I think their problem runs deeper than that. The problem is not that they have to migrate all these apps... the problem is they're using old methodologies (i.e., private contractors and vendors) to solve this issue. That's a proprietary way of doing things.

One of our readers, Shamar, had a brilliant and elegant solution: open the apps up to the community and let them take a shot at migrating the applications. (SteveOC articulated it even more succinctly with one term: munich.sourceforge.net. It seems a URL can be worth a thousand words, too.)

However you do it, it seems painfully obvious that the porting problem would be much less of a concern if many coders were to get in there and do the lion's share of the work. Would this be a simple thing to manage? Heavens no. But, this is a golden opportunity for the vendors involved in this migration to start an open source project that could (a) get their immediate task done, (b) create open applications for other government jurisdictions to use, and (c, the most crucial benefit of all) the marketable experience to go to future clients and not only say "we have open applications to use for your migration needs" but also "we have a proven migration management project structure that we will use to expedite your migration."

Could you imagine, having a ready-to-use migration plan that could be transported from government to government, or company to company, with only the need to just add developers? That would be a huge asset for vendors to have. And, as the set of ported applications grew bigger and bigger, the costs of migration for later government adopters would plummet.

Open sourcing applications would make migration a volume business.

Breaking old habits can be hard, I know, but in this case the benefits seem to outweigh the costs of sticking to the old models.

[Program Note: I will be taking Monday, October 31 off (no pagan rituals, I assure you) and leaving LT in the capable hands on Contributing Editor Rob Reilly. Catch you on Tuesday.]