Enterprise open source Toolkit

Story: What vendors really mean by 'open source'

• 
• 

Re: The big con - open source

: Open source is just a step up from shareware
: which really never took off.

This misunderstanding of the terms has already been covered, but it does kind of set the tone here.

: Over the years I have seen many companies
: waste millions of dollars investing in the latest
: technology only to see it either fail or not offer
: any real benefits over the systems they replaced.

Yup .. so have I. I have also seen that many of these investments were not the function of an internal decision on the part of the IT or commercial functions of the company concerned. The proprietary software vendors NEED their customers to keep paying them to write new code, so the vendors make it extraordinarily difficult for them to stick with what they already have. For an open source support company, a customer who's systems don't change for ages is a gift. The level of effort expended for their support pounds will be considerably lower than one that chops and changes with the latest greatest mandatory upgrades.

: Mark Taylor says that open source removes the
: 'Lock' down that exists in proprietary systems.
: What he is not actually telling you that there is
: really no 'Lock' down;

Proprietary file formats? To name but one.

: it is just the cost and effort to move to a new
: system. It would be exactly the same for a
: company that has been using an 'Open Source'
: solution for a number of years and wants to move
: to the greatest and latest system being sold to
: them.

If they are staying in the neighbourhood of the existing open source solution, then it is likely that only the broken bits will need to change. It is generally a disadvantage for proprietary software vendors to adhere strictly to open standards in file formats, protocols and such. It makes it easier for their customers to vote with their feet. This isn't an insult, it is sound business practice. I have argued that corner a number of times myself with management when I worked for such companies. It is a major advantage however for an open source tool to adhere to open standards. The chances are very high that they can then reuse whole libraries of stuff written for those standards. In the end case, the open source author will seldom put down any low level code at all and will simply bind together a bunch of libraries and put their own application code on top. This again was a fundamental difference I observed when moving from closed to open.

: Of course I see why Mark Taylor likes to bang the
: open source drum, his company is not about
: producing anything, but providing services on
: the back of other people’s efforts.

I would say that Mark's crew are in the business of producing functioning systems out of components that are generally open source. To put it in the rather derogatory way that you have is an insult to systems designers and product specialists everywhere .. open or closed.

: The 'Big Con', open source is being marketed
: by ‘Open Source’ companies that it is better
: quality then proprietary software.

This does seem to be, on the whole, the truth though.

: Strange it is the same people that have been
: developing the proprietary software that is now
: developing the open source software.

True, except that they are now developing open source using open source style development practices; mass code inspection, vast test groups, development tools produced using the same methods. I have developed in both worlds, when done properly, the differences are immense.

: There are thousands of open source projects that
: should have never been started, the quality and
: fit for purpose is non-existent.

These are the amoebic ancestors of the successful projects. Again the author is trying to fit closed source assumptions onto open source development techniques. If there is a project that died a death for some reason that was related to what I want to do, or indeed one that is still alive, I will have no bad feelings about ripping through it for something I can use. You can quite often save yourself months and months doing this. It is, after all, one of the basic reasons for open source's inception. Any project that "died" but who's parts were reused successfully in another was not a waste.

: The only companies making money out of open
: source (and there are many that are not) are
: service companies.

So you agree with Mark then?

: The problem is the open source market can
: support so many service based companies. So
: if open source software is to be successful it
: needs money to pay for continued development
: and support just like proprietary software or
: companies using it will find them selves in the
: same position when developers realise that they
: cannot keep working for free and the open
: source company disappears.

Wow .. a blizzard of bad assumptions.

You seem to assume that open source developers all work for companies that sell the software they write. Many do, but then many many do not. I have written both for sponsoring companies that wanted a particular function added, or bug fixed, for their own reasons .. and also for my own pleasure, or to relieve myself of a annoying bug. Some of this I was paid for, some paid for itself in reduced expenses in other areas (support etc). The assumption that all open source developers do so for free is badly out of whack I'm afraid.

Another assumption drawn from the closed source world is that the company MUST keep slamming out new code for it's own survival. The opposite is true in fact. Once the code has reached a level of maturity that it can stop being hacked about and settle out the last of the deep bugs it no longer needs much if any development attention. Open source encourages cooperation and reuse by it's very nature.

Bear in mind that as the general level of reuse goes up, less code has to be written. You are already in the strange position of having the developers from 3 or more implacable competitors sitting down together to design and develop the code their companies are going to sell. You better believe the heat of competition leads to excellent code being written. This of course instantly drops the development bills for all 3 and longer term leads to fewer problems due to the white hot zeal of the competing developers. If you have not been in this position, you simply would not understand the marked difference between the two scenarios. It took me a while to click and I was sat there with them.

: As my father kept telling me as I grew up I need
: to get out there and get a paid job as there is no
: free lunch. Sooner or later open source software
: companies will have to do the same one way or
: another.

It is plain that you have taken core assumptions from knowledge and experience gained in the classic closed source proprietary software market and transferred them, unchanged into the open source market and are now using those assumptions to "prove" that the open source market is doomed. Your case is further helped by those people that physically did this and opened an open source company based on closed source principles. They failed, or in the better cases didn't do well until they "got it".

The open source market is way more "free market". it has to be. There is nothing stopping a new company picking up the entire software stack from another company and setting themselves up in competition. The most famous example being Mandrake, way back when, when they forked the Red Hat product and came out with a less buggy and better supported alternative. They made money out of that. They also encouraged their donors to pick their act up a bit. Everybody wins.

• 
• 

• 
• 
• 

Full Talkback thread