Linux During The .Com Crash

freakboy303 writes "ZDNet has a short article that can be found here , It basically talks about what the last couple of year of gloom and doom mean for the linux world in general. It seems to me it would make it more appealing to .coms to use the free software but..."

Comments

Over maybe in the investment sense?

[syrupMatt]

What the article seems to push at (albiet around the bush) is that there are less companies willing to stake their future on the sales of service for Open Source work. Although RedHat and a few others are posting profits, the overall tech downturn is probably preventing any speculation in o.s. based companies.

I think the point is missed however, if this article is taken as a view of an overall decline in open source work. If anything, now is the time for developers to be able to work at a less pressured pace, since they aren't worried about advancing the project so that Company X doesn't go out of business before it can put together a viable distribution/product/release.

.com crash perfect for Linux

[toupsie]

The main problem I saw with the ".com's" I worked with was that they bought the most expensive servers (i.e., Sun, DEC & microsoft) they could get their hands on from the get go. They just figured they would have 1,000,000+ visitors a day and equiped for it. I am sure it impressed the VC suits as well to see their invested cash going for "quality" hardware and Operating Systems.

In reality, these ".com's" should have taken off the shelf hardware from CompUSA, fdisked the harddrive, popped in a floppy and FTP installed Linux or BSD. Once they realized that the load was more than the servers could handle then they could have thrown money at the big iron or betting yet, just add on more Linux/BSD servers and scaled up.

Its no wonder that Sun is on the skids right now. You can get barely used, high end Sun servers for pennies on the dollar in the 2nd hand market. I just saw Sun E250s being sold for $1750 today that were $15,000 a year and half ago. Not a bad deal for the user, a major disaster for Sun.

Re:.com crash perfect for Linux

[SuiteSisterMary]

The main problem I saw with the ".com's" I worked with was that they bought the most expensive servers (i.e., Sun, DEC & microsoft) they could get their hands on from the get go

That's really kind of a 'flip a coin' problem. I'm a firm believer in getting as much as you can up front, because upgrading is a real bitch, both in terms of getting what you need, and the actual downtime/replace/blah blah blah bit. Also, you just never knew when the shit would take off. We had that problem; a prototype gets pressed into service; sure, limited to a few customers. Then, suddenly, everybody's trying to use it, you're struggling to build your PRODUCTION system to handle it, customers get tired of knocking on the door and leave for a lesser, but available, product, and you get blamed every time the 'producto-type' goes down.

Re:.com crash perfect for Linux

[ergo98]

What? I'm not supposed to have a chair?

While all the attention is put on those chairs as some symbol of .COM excess, there were far worse excesses. Something as simple as unnecessary, extravagant travel by senior members. One unnecessary "business trip" by a CEO can be about a dozen ergo chairs. One CEO making $7,000,000 a year is quite a few thousand super duper chairs. I just find it odd that everyone jealously, it seems, focuses on those damn .COM workers with their Aerons when so much ridiculous excess happens daily in the corporate world.

Open source programming is like playing solitaire

[iabervon]

There seems to be all this concern about whether people will write software if they derive no obvious benefits from it. This is all based on the misconception that people dislike writing software. Many of the same people who don't understand will play solitaire when they don't have to (and even when they're not supposed to). They derive no obvious benefit from it, nobody cares how they do, nobody pays them, and the damn thing doesn't even stay solved.

Writing OSS is like playing solitaire, in that it is fun (you're solving little puzzles which are non-trivial, but not impossible), but when you've done it, you end up with a program that does what you like, and you can give it to people and they'll be impressed. Some people might even pay you. Of course, at some point they start expecting you do what they want rather than just what you feel like.

People get paid a huge amount of money to play basketball. Other people don't even get reimbursed for buying a ball and a net, but they play anyway. The same thing is true of writing software.

The frustrating thing. . .

[foo fighter]

. . .about surveys of Linux usage in business, is that they are all to frequently based on "spending priorities for executives" and "new license revenue shipments". At least this article mentions that linux being available for free will skew the results in the proprietary offering's favor.

I am trying to sell my boss on bringing linux into our educational institution, both on the desktop and on our servers. When I show him and our CFO that upgrading all of our desktops to Windows 2000 will cost us $100,000 up front while Linux is free they get excited. But when they see reports that only 2% of shipping desktops come with Linux they get understandably (seeing it from their POV) concerned.

It would be nice to see a metric like "Six of the most popular linux distributions report sales of 100 million units, and downloads 500 million units for fiscal year 2001" from organizations like IDC and Gartner Group. That would help account for sales AND downloads and hopefully skew the numbers back to a more correct figure.

Of course there is still the problem of counting installations after the initial purchase or download. Any number you get will be much fuzzier than the "sales and downloads" figure. The solution is to survery the engineers and not the executives. Ask the engineers how many machines they installed their copy of linux on and you will get a much more reliable figure.

The most interesting thing about this article is the problem of linux competing with pirated Microsoft software in third world countries and southeast asia. In these places Windows is effectively as "free" as Linux in monetary terms. When all you care about is price parity, why not choose the more popular of the free solutions?