Open source

[Richard Conyard]

Just Friday afternoon ponderings... but can anyone tell me the business case for releasing open source software?

That is, why I would choose to as a business write a piece of software and make it open source, rather than using open source (which we've a fair smattering around the office).

[Top Hat]

We're currently working on a script, when its finished, we will release it, open source for others to use.

The Reason: selling software is not what we do. we will ask all those that use it to link to us or join our affiliate program, we will also put it on hot scripts etc. So we may get some inward links.

[Richard Conyard]

I guess, so small projects the good will publicity would pay for the development time; I can see that - it becomes a marketing exercise.

Larger, more time consuming pieces of work?

[Top Hat]

Well the script is for us, but once done we may as well release it, we may get some inward links

Shall I tell you what it does?

[Richard Conyard]

Got for it :)

[Top Hat]

Oh, I am glad you asked :D

Its a script that add a (c) message to any hot linked images, It has a whitelist of images and sites that can hotlink, a blacklist of sites who can not hot link (they get nothing) all the rest get an image with a copyright message

e.g.
http://www.sillyjokes.co.uk/images/dress-up/acc/hats/historic/tophat-black-felt.jpg

Look at this image on an approved site you get no (c) message.

And it counts, and list the biggest hotlinkers

[Enigma121]

I guess, so small projects the good will publicity would pay for the development time; I can see that - it becomes a marketing exercise.

Larger, more time consuming pieces of work?

Large software projects are very, very difficult to produce to high quality standards. If we take perhaps an operating system as an example.

A company the size of Microsoft still manages to produce software with many bugs through the process of "closed shop" development.

In the open source software model, bugs are openly revealed and get quickly fixed by the end powerusers. This means that overall the standard of software is very high and costs are minimal. We see this in the Linux community.

Trouble is that someone, somewhere needs to make money on this to keep it viable. Hence many opensource developments move into "commercial license" arrangements when you come to look for support.

Linux distros have also introduced the "license trap" whereby they wait until a large enough commercial user base is dependant on a particular product, then make it a chargable entity that the user will then struggle to do without.

Safe to say, that a savvy open source company has no trouble extracting cash when the time is right. In many cases, they are extracting cash for stuff that is already public domain. It's all about bundling of software, solid QA procedures and good marketing.

[Richard Conyard]

Nice work TopHat :-)

Thanks for the comments Enigma, but I still don't have enough ammo to suggest a change in our current plan.

I guess I'd better come clean about the reasoning for this post. I am weighing up the benefits of making our sxForms open source, and I need to justify it on business grounds.

sxForms are our form rendering engine ( http://www.thinkcolony.com/whatiscolony/technical_details.asp ). They are exceptionally powerful, but with more development time than we've got available could be a lot more so.

If I open them out to open source I potentially get:
A larger dev community to work on new controls
New render engines
Greater awareness of the Colony product (since it's 100% sxForms)

But I also get:
The ability for our competitors to use the technology
Lose the cutting edge since we're giving it away

It's a tricky one, I want the benefits without the cons and you can't do it.

[james.hill]

I see the reasons for open source to be as follows:

You sell a complementary product (IBM)
You sell complementary support/documentation (SQL-Ledger)
You sell a packaged corporation friendly version of the product (snort (I forget what the commercial side of it's called))
You're business depends on it, and the benefits of sharing the code outweigh the commercial benefits (ISP software such as bind, exim, etc).

For your specific case, I would say that the main advantage is the increased awareness of your colony product. You can host the project on your own website with links pointing wherever and push your own product.

You summed up the negative pretty well.

It's worthwhile pointing out that if you want to keep the cutting edge it's perfectly possible to release your older code under the GPL and retain your own control of the latest code. SugarCRM would be an example of this in action.

[Enigma121]

If I open them out to open source I potentially get:
A larger dev community to work on new controls
New render engines
Greater awareness of the Colony product (since it's 100% sxForms)

But I also get:
The ability for our competitors to use the technology
Lose the cutting edge since we're giving it away

Looking over Sourceforge, there are already a lot of CMS systems in open source. This means that your competitors already have a stack of code to start with.

It's possible to release under more restrictive licenses which allow mods and usage of the software but prevent production of derivative works. Of course you always have the battle of proving this has happened.

You will frequently retain big corporate customers even if you go open source, as they like to have licenses and the assurance of support and someone to kick if it all goes wrong.

If I were in your shoes, I'd probably go open source to remain the market leader.

Apache is afterall doing a bit better than IIS when it comes to deployed web servers. Think of the prestige and positive traffic as a result. You need to have a commercial offering ready to catch the big corporate customers that will be drawn to you.

[Richard Conyard]

The difference is Engima the people who put the hard work into apache and apache ssl are not seeing the financial benefits that they would if they madea a charge for it.

From a technical case I'd love the software to get wide adoption (it's not the cms, just part of it), but that doesn't really hold up the business case to my MD who can quite rightly turn around and say "so I'm spent XX man months developing this system and you want to give it away??!?"

[Rob Holmes]

Richard - could you use that Open source release as a hook then upsell to your other packages?

Rob

[Richard Conyard]

Actually in the process of that with another one of my tools dbsite - it's awaiting approval over at sourceforge. Here's the description if anyone is interested:
DBSite: Written in Visual Basic this DLL helps solve issues for ASP web developers, e.g. handling file uploads, safe parameter population for stored procedures (sql injection checks), SQL XML calls, file system browsing, etc.

[Richard Conyard]

It's been approved :-)

Link by 'ere: https://sourceforge.net/projects/dbsitedll/

[james.hill]

It's been approved :-)

Link by 'ere: https://sourceforge.net/projects/dbsitedll/

Congrats =D

[Richard Conyard]

First files posted - although they still need a tidy up.

[Top Hat]

The 'image hot link ' script that I was talking about is now availiable. more info here

http://www.ukbusinessforums.co.uk/viewtopic.php?t=6692

[Richard Conyard]

http://sourceforge.net/projects/dbviewasp/

And another one - the source code is awful, but the tool is useful. Been using it on and off for about 7 years now in one shape or another.

[mattk]

Richard - just because your software is open-source, doesn't mean you will have a large development community. No offence, but why would someone write a mod for an obscure CMS system when they can instead be writing hacks for Firefox, Bittorrent etc?

Secondly, just because someone in the open-source community writes a mod for your software, it doesn't mean that mod will:

a) work
b) be well written
c) comply with your architectural design.

I use open source software for certain tasks and I have even modified it to meet my needs. However, my modifications could never be used by the original development community, mainly for the reasons above.

[Richard Conyard]

Matt,
I know what you're saying, but it doesn't seem to fit in with the thread except by the tenuious link of open source.

To be honest I don't think we will ever open source the CMS. There has been too much hard work gone into it (and we're generating too much revenue out of it), for there to be a compelling arguement to start giving it away free.

Technologies associated with the CMS though might be an avenue to go down (thus starting this thread), however it does open up the trouble of parting with some technical advantage.

With the to links I've posted up basically it's old stuff that whilst useful no doubt to small dev / design houses doesn't really give anything away so no business impairment there.

[mattk]

Richard - don't equate open source to mean free. I use a piece of software for issue management, called Gemini (http://www.countersoft.com/), it's open source, so I can make changes if I need to, but it's certainly not free, the commerical licence is £149.

[Richard Conyard]

Still I would feel uncomfortable putting the CMS out there as open source. We do have open APIs, but the source remains closed.

[mattk]

What language do you use?

[Richard Conyard]

Some VB DLLs, some C and some ASP.

The ASP obviously people can view the source of, however if they modify without permission it causes problems with support, so that's a no no.