[OT] Open Source and my company's web application
Sy Ali
sy1234-Re5JQEeQqe8AvxtiuMwx3w at public.gmane.org
Fri Jan 20 02:34:04 UTC 2006
First, thanks everyone for feedback. I've already started the effort
to figure out just what code has been shamelessly adopted, then I'll
go check out licenses while we figure out just how much code has been
modified at our end. After that, I want to get patches delivered for
any changes we made, so that things are cleaner. I think all of that
is necessary regardless of any decisions on the project's direction.
Aaron mentioned the "always on" aspect. Soon enough I will want
database mirroring to another server (a fallback server). This has
already been researched and is doable. MySQL has some nifty features
for having secondary servers. After that I'll want to have a
mirroring function so that I can keep a copy on my machine.
It's been a good experience to have things online like this.. now my
biggest worry has to be the office net connection.. it's "low
highspeed" to be polite about it. Maybe I should start a separate
thread to whine about that. ;)
At some point it'd be nice to have the application usable offline on a
computer and maybe one day (probably never) on a portable toy of some
kind. Being able to use it offline and then connect and have it sync
the changes would be nice.. probably hairy with good old wiki-style
"edit collisions" but maybe it can be done.
Now as far as "trust" goes.. I completely agree that there are trust
issues. I don't like the idea of having my most irrelevant todo stuff
hosted elsewhere, especially if I can't easily download it and use it
on my end. I certainly wouldn't have business info hosted somewhere
shared. The idea that a business might want a chunk of a server set
aside just for their purposes is an intriguing one.
I think that some sort of publically-viewable thingy has to happen,
regardless of the code being open or not. Maybe that first step will
get feedback that helps us with our decision. I kindof want some
engine by which all the feature requests and bug reports for it are
open.. because that will demonstrate some of its functionality. Then
the rest can just be a prepared demo and a sandbox on the side.
No we're not into selling services, but we're starting to look in that
direction. We've been asked to set up hosting for a few customers and
that's gone well enough that we're perfectly fine with reselling that
service. Applications look almost like free money.. carve out a piece
of a host, set up a free application, tweak it and support it for a
customer.
A strange thought.. if the application were completely GPLed.. but we
were a major contributor of code and insight, I think that alone would
really favour us for providing hosting or customization as a service.
In this sense, our continued interest in it would tether it fairly
close to its old home.
Again, some of this is dreaming.. we still have it in mind that it
would require a complete rewrite from scratch at some point. Sure it
looks good and it works quite well.. but the real world is much more
harsh than a developer's box. I know there are a lot of
creepy-crawley features waiting to be found. =)
I mean.. the real world would want cool things like .. skinnability,
easy multilingualism, administrator-controlled featuresets and all
kinds of wonderful things which aren't making us itchy right now. =)
As an aside, I think the biggest thing I'd want changed in a rewrite
is to use postgres instead of mysql.
--
The Toronto Linux Users Group. Meetings: http://tlug.ss.org
TLUG requests: Linux topics, No HTML, wrap text below 80 columns
How to UNSUBSCRIBE: http://tlug.ss.org/subscribe.shtml
More information about the Legacy
mailing list