Once in a while the call for open sourcing of Notes and Domino comes up. Actually I personally would like that. But would it really boost the use of Notes and Domino? That is the point were I have doubts.
Let’s just forget for a moment some issues that will prevent this, because apparently there is some third-party code in there, IBM can not just give away. That’s a legal problem.
Let’s focus on the other problems, which are huge enough.
Technical
While I have no doubt, that there is a great many code that is wonderful, we had some snags in the not so recent past, that annoyed some of the best yellow bleeders so much, that they recoded a whole bunch of things, just to have something that works as it should (I just say Java).
When we would dive in the source, we would probably find quite a few more things, that should be rectified and that does not pay the rent.
While Notes is based on a Non-SQL-Document-Database, which are very en vogue today, the performance of the NSF is not at par with MongoDB or CouchBase (or CouchDB for the matter). Replacing NSF with CouchDB would give a huge performance boost, but that would demand quite a bit of new code. Some logic could certainly be reused, but still, somebody would have to write a few thousand lines in Erlang.
A few other thing were to address, the lousy performance on the Mac, search isn’t that good either in comparison. I think some of us would gladly kill the Eclipse RPC.
Domino becomes more and more expensive to maintain, too. A whole bunch of new code would be required to make administration easy and save. Web Admin comes to mind, too. There isn’t a way around a web admin.
And let’s not forget the slight security problem we have with it. Not addressing this, would be certain death of the whole thing.
We would have to do without Watson, that wouldn’t be a show stopper, but probably we would have to live without Traveller. That would be a show stopper. Developing a Traveler replacement would be quite a task.
Repeat after me: „Migration must be painless“ and let’s not forget updating it should not be rip and replace.
And to round things up; to be a success, it can’t be „the same old“. Feature- and performance wise it would have to be something out of this world.
Design
The one most important thing I hate about Open Office (and Libre Office); it’s ugly. I hate to look at it. It might be technically good, but I still hate it. Notes hasn’t very compelling looks either, but still better. But to be a success, that Open Source Version 1.0 would have to be the best looking email client ever. As an example we have only to look at Android. Only because Google and others have invested in good Design, it became a success. For us geeks, Linux was always the better server choice. But as a desktop, it never really succeed. I think it is because of the desktop design. It just isn’t up to specs. Or better, it never was overwhelmingly better as Windows. Only then it would have seen as a replacement. People hate change.
But who is going to do the design? A real good designer with a vision would be required. John Ive isn’t available and I do not believe in hundreds of designers fiddling around with some design language. For a single product it has to be one person that makes all the design decisions.
Organisation
While the Apache Foundation does one hell of a job, most of the really important tools are not made for the run of the mill end-user. I doubt, that Apache would be the best way to organise the work. Notes would deserve a dedicated organisation and it probably would need one that works full-time. Therefore it has to make money. Who pays?
Probably Linux could work as a model. Having a few experts who decide where to go, is probably better, than having a democratic vote. These half gods would also be responsible for preventing any attempt of forking.
The Linux model also has the advantage, that it lets companies make money with it.
For our purpose, the initial financing could be done by the highly committed customers. Getting them in the boat and building a project and team based on Kelly Johnson’s 14 principles, is the way to success and fast.
Probably we should only focus on the server anyway. Clients could be the thing to make money. Except for basic client that serves as example and first building block, if a rich client is really needed.
Mobile is without question one of the corner stones that must be included.
Marketing
Who is going to sell that thing to companies who are absolutely against open source software for their strategic tools? Or the ones that are fed up with Notes anyway? Or those who have only heard terrible things about Notes? Only keeping the few companies that are willing to dedicate time and resources to an open source project, isn’t a viable solution for all the partners and the product. And there is the fear about migration. That should be painless. I mean really. Not just a sales pitch.
Small companies might want to go for a free version, but please in the cloud.
There are tons of ideas, how to build a business model around this and one would certainly be a good one. But is a group of volunteers capable of doing this or will there be an eternal fight between those who want to feed their kids and those who rather starve than going against the higher ideals of open source?
I believe, that IBM rather lets Notes/Domino die, than give it away. But if IBM does it and it becomes a real success and IBM would have to explain that to the greedy share holders.
Having said all that, I believe the better solution would be starting from scratch altogether.
Getting the code and finding out, that more than half of it would have to be rewritten anyway, would be a bad surprise. Starting with a clean sheet gives the opportunity to get the best ideas from the best people and make it happen much faster.
From the start the group could build something that really can change the game. For example a server, that does not care which client it serves. A server that does more than just mail, also chat, SMS and documents. A server that uses Apache as a web server (but carefull: the more third-party code, the more dependencies).
A structure that lets one store anything project AND contact based.
Thinking about it, it should be something like Connections for the poor, only with more features and fewer servers.
Still, it would be a lot of work. But fun. I would help. But frankly, the yellow bleeders are not the open source geeks who flip burgers during the day and do miracles during the night. Somebody would still have to come up with a sustainable organisation that keeps the project going and contributors happy. Even if that means the final product will not be free to use for all.
But since we know by now all the remaining Notes customers, all of them could tell us what they want from the new thing not called Notes at all. That’s a plus.
Oh, and migration should be painless. I did mention it, did I?