Open Source?

So, Informix is to become two operating companies under a single holding company? Interesting! Pity about the 3Q2000 earnings warning. How many databases is the Informix Software (database operations) company going to be looking after? SE, OnLine, IDS, XPS, Foundation, RedBrick, Cloudscape(?), Universe, Unidata, ... If you were in charge, would you make any of these products available as Open Source products? Which Ones? Why? Or Why Not? Would you try to join the development team? Would you use the Open Source version yourself? -- And what is the Riddle of the Sphinx? Sent via Deja.com http://www.deja.com/ Before you buy.

The Great Sphinx <sphinx_101@my-deja.com> wrote: > [...] > How many databases is the Informix Software (database operations) > company going to be looking after? SE, OnLine, IDS, XPS, Foundation, > RedBrick, Cloudscape(?), Universe, Unidata, ... > > If you were in charge, would you make any of these products available > as Open Source products? Which Ones? Why? Or Why Not? Would you > try to join the development team? Would you use the Open Source > version yourself? I am who I am. I do not work, directly or indirectly, for Oracle. I do not see that asking this question qualifies me for retorts such as 'Sphincter' and 'A-hole'. Judging from the response thus far, one person would be interested in using Open Source versions of the Informix databases, but no-one can be lured into helping with such an effort. And it isn't even interesting to speculate on whether it could happen? OK, that's fair enough. Dreadfully disappointing, but fair enough. Have we reached a complete consensus on this, or have people decided not to answer simply because the person asking the question chooses to sound...but both Delphi and Oracle have the wrong connotations; have people chosen not to answer because the questioner chooses to remain unidentified, and hence people have assumed that the question cannot possibly be serious? Why are Obnoxious Clowns and Mindless Blonds acceptable but innocent Sphinxes outcast? And where is the Mindless Blond these days? There's been an interesting amount of discussion about my signature. -- And what is the Riddle of the Sphinx? Sent via Deja.com http://www.deja.com/ Before you buy.

In article <8qdohb$in2$1@nnrp1.deja.com>, sphinx_101@my-deja.com says... > The Great Sphinx <sphinx_101@my-deja.com> wrote: [...] > Judging from the response thus far, one person would be interested in > using Open Source versions of the Informix databases, but no-one can be > lured into helping with such an effort. And it isn't even interesting > to speculate on whether it could happen? OK, that's fair enough. > Dreadfully disappointing, but fair enough. Not sure if you like to hear this, yes it is Informix product, and yes, I think it is at least the 50% of the reason Informix exists as company, but it is not a database. It is GNU 4gl called Aubit 4gl, Informix-4gl(TM) compatible 4gl compiler, but actually quite more than that. It was originally developed by Mike Aubury of Aubit Ltd. and is now being developed as GNU/OpenSource project at www.falout.com/projects/Aubit4GL Not that we where screaming and shouting "come join us" (we where not ready to provide all the services this kind of project requires - we are much closer now), but judging by number of downloads and visits, and number of people that offered help, I would say that you should not have too much hope for even more sophisticated software like database. No, that is not really what I wanted to say: I am the member of several big GNU development teams, and my assessment is that COMMERCIAL COMPANIES are driving sophisticated project, like Apache, PHP, etc. Sure, they did not started that way, but what happened in Linux world is that first small companies spotted the value in this kind of cooperation, providing support and services for the project, as well as some key people in development, that where on the payroll of this companies all the time working on GNU projects, that companies (mostly small ISP's and web space providers) then used in there commercial operations. Then some bigger companies like IBM joined in, and here we are. This kind of projects, especially if we are talking about database, require solid leadership, vision, planing, and most of all, dedication. Sure, you can have bunch of people coming and going that occasionaly fix some bugs, but if you don't have REAL developers that build strategic features, you will not have any bugs to fix in the first place. So to summarise, to have successful OpenSource project of database kind, you will need a sponsor, a company that will have the vision of benefiting form this kind of project, and be willing to provide one or two EXPERTS in the field to do more or less full time work on this project, and still receive salary. If you can find this kind of people, and make sure they will be able to stay on this project long enough (and we know this kind of people do not come cheap), you can have successful OpenSource project. If not, you will depend on people of varying level of knowledge with varying level of available time, where X is typically close to zero. If you do succeeded in putting together this core team, people will join, debug, write documentation, add smaller features, report bugs, discuss issues, etc. Aubit4gl is still looking for this core, and I'm trying to find company that is willing to pay for Mike Aubury's time. However, I will be announcing the project official soon, more to let people know there is a 4gl compiler that is in it's maturity approaching commercial usability, then to expect some rush of developers that will be joining us. Not that we would mind, but I don't think this will happen without what I described. I do not believe Informix would have the vision even to release code for SE, but if they would, I'm almost sure they do not have the vision to dedicate at least one expert and one administrator to work full time on the project. But if they do release any code, what someone would need to do in my opinion, is to find a company with vision and resourves to provide development core for the project, in anticipation that results will be useful for they day to day operations, and could more then pay for the effort. Additionally, having some knowledge on Informix databases, I don't think that anything more sophisticated then SE or perhaps old OnLine have much chance to be developed in this way, because of level of knowledge required even to just use it properly, not to mention develop it. Current Informix engines are operating systems in itself, and in some respects more sophisticated then many operating systems on the market today. I do not think more then 100-200 people on this planet shoud even discuss possibility to contribute anything to such beast, other then possibly fixing bugs and doing OpenSource project related work other then development and design. Myself included. This 100-200 people command 300-500.000 US$ a year (and more), and I do not think they will just let this money go, and dedicate there time to honour and glory. You need company willing to pay this to get this code working for there business. So you need company that is paying significantly more then that currently for there database needs. Or you need miracle. Miracles in OpenSource community do happen, but you still need a reason for them to happen. With PostgreSQL already a very solid database, SE will not be interesting to many people. With IDS 7.x released to OpenSource, you can revolutionize the industry, but this would force Informix to transform into SERVICES based company. Given the fact that they where never good in selling PRODUCTS in the first place, this would be an excellent idea in my opinion, but I don't think Informix was capable of this kind of vision in last 15 years. -- Yours, Andrej Falout, http://www.falout.com ICQ 7628616 ++64.21.607517 #----------------------------------------------------------------- globals "std_disclaimer.4gl" Ask yourself just one question: ' Qu' m's se puede hacer y aprender ? - Propellerhead ReBirth RB-338 manual

I read Andrej Falout's reply, but am replying here in the interest of keeping this post a bit shorter in length for readability. -- Andrej wrote: "So to summarise, to have successful OpenSource project of database kind, you will need a sponsor, a company that will have the vision of benefiting form this kind of project, and be willing to provide one or two EXPERTS in the field to do more or less full time work on this project, and still receive salary." -- Good point and I would tend to agree. The company with the software needs to take it serious, and make sure they COOPERATE with the open source community. I won't say any more other than the last go-around with the Linux port was a complete disaster from a marketing perspective with developers, even though, yes, it had some highpoints, and a complete success in pissing everyone off. It made everyone less than willing to really work together because there was no incentive on either side to take things to the levels necessary for success. The Linux port was a great education in random acts of completed projects. It turned out to be a miracle in the end. In this business, MOMENTUM is the key, not just occasional excitement. Ask Oracle and Microsoft and IBM. I had heard a while back that C-ISAM was a candidate for open sores, but it didn't go anywhere, this was back in February. Why have an Open Source project? What is the real point? Increased mindshare? Increased Marketshare? What might be better than just an open source project is a developer program that allows Informix to be competitive with co-op programs for developers. It falls under the affiliation-style marketing concepts many websites and dot-com companies are using to push product. There are a lot of ways to make this happen. If there were more developers aware of Informix and the benefits, and they were rewarded accordingly, I think that open source would stand a better chance of happening with lower risk to Informix. The hook for Informix: they get increased presence in the market without having to hire more staff sales people. They already have a solid following of a small amount of developers, what these developers really need is an incentive to help on not only open source, but on REGIONAL sales. I presented this idea back in February at a meeting with a now departed exec, and it was favorably received. The other benefit for Informix: they essentially can sell the software for whatever price they want per customer, and concentrate more on the professional services side, charging for the consulting by bringing the REGIONAL independent consultant into the sales program, and allowing a VERY competitive solution instead of being forced to sell software at fixed prices. Very scalable. For the developer it's like another level along with certified DBA and such. It would allow an instant increase in a regional sales force without having to expand headcount in regional sales offices. By bringing in the independent consultant into the sales cycle, you effectively and realistically reward the very people who are more than willing to promote Informix products. And this includes the loyal following of the consultants working with Ardent and RedBrick. From a developer standpoint it is a real working partnership with Informix in a "Think Global Act Local" kind of way, with real CASH incentives. It would allow regional sales offices to build traffic and concentrate on regional developers in meaningful ways that are appropriate to a particular market. The region would determine marketing goals based on building developer programs in a region appropriate for that region, and of course with a few goals from the home office. The regional sales people would also not have to worry so much about having to know everything about all the products, what they could concentrate on instead is making their customers feel special, and important. This would expand market/mind share faster than anything. Everybody would win, and of course, developers being the way they are might even offer up some of their time in an open source project knowing they would be part of something meaningful. Open source only works when both sides are open. Think about it... Informix is new again, and hopefully the showstopping arrogance and ignorance has cleared out of the building. Tim As a reference to some of what I'm talking about: http://www.affiliatefind.com/ ( Note: this is only part of the picture, think outside the box! ) The Great Sphinx wrote: > > So, Informix is to become two operating companies under a single holding > company? Interesting! Pity about the 3Q2000 earnings warning. > > How many databases is the Informix Software (database operations) > company going to be looking after? SE, OnLine, IDS, XPS, Foundation, > RedBrick, Cloudscape(?), Universe, Unidata, ... > > If you were in charge, would you make any of these products available as > Open Source products? Which Ones? Why? Or Why Not? Would you try to > join the development team? Would you use the Open Source version > yourself? > > -- > And what is the Riddle of the Sphinx? > > Sent via Deja.com http://www.deja.com/ > Before you buy. -- . .- .-- .--- .---- Tim Schaefer .----- tschaefe@bellsouth.net .---- http://www.inxutil.com .--- .-- .- .

In article <39CB4E4C.86297059@bellsouth.net>, tschaefe@bellsouth.net says... [...] > In this business, MOMENTUM is the key, not just occasional excitement. > Ask Oracle and Microsoft and IBM. Absolutely, and it can be sustained only with some serious and persistent support from stabile base. It can be Informix itself, it can be some clever Informix customer, but it must be there. It may become redundant after some time, but only very successful OpenSource projects can achieve that after years of MOMENTUM based on this kind of base. [...] > Why have an Open Source project? What is the real point? Increased > mindshare? Increased Marketshare? Both. Time=Money. Money=Time. And I would like to add, it is not, or it will soon not be, an option, but necessity. In my current project, all significant components, with exception of Informix engines and D4GL/4Js CLI-Java are OpenSource (Linux, Apache, JServ, JVM, SSL, PHP...) Guess what components are slowing us down in development? The ones without SOURCE CODE. We made more then few additions and changes on most of the OpenSource components, to accommodatee the need that we had to put this thing together (ASP Java based sistem running ERP type applications written in Informix 4gl), or just plane bug fixes, in the process contributing to this OpenSource projects. With closed source, I can only report issues and pray (R&P). No matter how much resources I have here, and no matter how much knowledge, I cannot do anything more. I would, and I could, but I can't. Anyone familiar with ASP idea, and it's curent attack on the market would instantly think of TIME TO MARKET thing. Ring any bells? This is no longer an option. In the future, wherever time to market have to do anything with the projects I'm managing, closed source ANYTHING is out, out, out! If at all possible. I will use PostgreSQL engine not to save money, or because it is better, but because I get the code. Wherever possible. If I can get the code, with accompanying OpenSource development project supported by Informix, I would be prepared to pay EVEN MORE for licenses and/or commercial production support to Informix. Same with all other tools I need. I'm saying this just to clarify that my motives have nothing to do with money. And on top of that, this is only the one of many advantages OpenSource model brings, it just happens to be most important to me at this moment. But there are many, many more. So I think, this is no longer an option, it is a necessity. Traditional turnaround in tools development and maintainance is simply not acceptable, when you must compete on the market the other guy can do thing twice as fast. This is the main reason I give my time to Aubit4gl. I believe in language, and still did not find more productive tool for business related database applications. With a little luck, I will soon be able to replace Informix-4gl in my projects, and since Aubit4gl can also use ODBC, I can replace Informix database engines. OK, Informix now have D4GL that can also use ODBC, and I do like and use the product, and I also like and use there engines. But to be very honest, my project was in danger several times, just because I was not able to show RESULTS in TIME TO MARKET frame needed by my client, and only because resolution on problems we had in this relatively new field was TOO SLOW. Not that they did not try, and not that they did not do things for us, but the whole traditional mechanism is limited by it's design. And I definitely don't like when I work on the project for a year to have fears that it will be canceled because I can't get solutions in time needed by my client. If my project have to fail, I prefer to have it fail because of my inability, or lack of resources. This are the thing I can do something about. Not because I'm using tools that cannot compete with market. [...snip of description of business model...] Yes, this sounds like good plan. Two things I would like to emphasize: 1) What you just described, is actually the model many application vendors where using for years; for instance, company I work for at the moment, that have ERP product based on 4gl is doing more or less the same thing with there ERP application and code. Most of this is based on availability of the source code to the distributors/consultants on the field. Application vendor is actually work as core OpenSource development team here, with Open in Source being Open to limited number of external people, but the model is same. What Informix would do in such a plan is to provide the ability for application vendors to act just like there distributors/consultants do on application level, just on tools level. 2) Extremely important in my view is the distinction between PRODUCT based and SERVICES based business model. Software in this scenario is no more then building block for the PRODUCT (from the customer's point of view) and what is the value here are the SERVICES both the application vendor and distributors/consultants on the field can in this scenario provide, that would be impossible without the access to the source code, and/or support from the tools vendor that is SERVICES oriented, and have OpenSource development model. Yes, database is a tool after all. And yes, you need a SERVICE to put it to actual use. How many times did I hear "we would like to use Informix products, if only we can find people with knowledge to provide SERVICES to us. But, we can find Oracle people everywhere. Sorry." -- Yours, Andrej Falout, http://www.falout.com ICQ 7628616 ++64.21.607517 #----------------------------------------------------------------- globals "std_disclaimer.4gl" Ask yourself just one question: ' Qu' m's se puede hacer y aprender ? - Propellerhead ReBirth RB-338 manual