[schooltool] More background on Janastu

[Ronald Wertlen]

PS Sorry about any confusion from the last post (bit rushed). More
clarity/info on my blog...

http://ekhayaict.com/wordpress/?p=78

[Tom Hoffman]

Post by Ronald Wertlen
PS Sorry about any confusion from the last post (bit rushed). More
clarity/info on my blog...
http://ekhayaict.com/wordpress/?p=78

Hi Ronald,

I flat out missed this email, and I still can't find your "last post."
So sorry for the slow response.

I don't think there is much we can do at this point. For better or
worse we are tied to Zope 3, and our codebase is rather abstract and
sophisticated. SchoolTool started with very ambitious objectives to
create an extensible, i18n and L10n-able framework for schools around
the world. This is enterprise software.

The problem is that right now the value proposition for a project like
Janastu to use SchoolTool is at a low point. The codebase is complex,
individual components work and are being tested and used in a handful
of schools, but we are still in the process of pulling this together
into a coherent suite.

So for web developers without heavy object oriented and component
design experience, getting up to speed on SchoolTool is a big step for
an uncertain benefit. It is also an uncertain benefit on our side.
We have a pair of developers. If they're spending time on IRC the
next six months trying to tutor programmers on the fine points of
ZCML, they aren't finishing SchoolTool 1.0 beta (due in October).

There is a long term plan for solving this problem. First off, more
complete releases of SchoolTool make its value more certain.
Secondly, I'm hoping Grok 1.0 http://grok.zope.org/) and SchoolTool
1.0 come out around the same time (i.e., next April), and we'll be
able to start leveraging the ease of development Grok brings to Zope 3
for people looking to extend and modify SchoolTool.

In the meantime, however, our priority is to keep our heads down and
finish SchoolTool.

Having said all that, I'm always happy to offer advice and feedback to
any open source SIS or MIS project.

--Tom

[Ronald Wertlen]

Hi Tom,

thanks for the response.

I have looked into the code of Janastu (what is available on the Google
svn) and they are obviously also using a framework. Additionally, they
are tied to a legacy CMS (Pantoto) which the commercial outfit behind
the non-profit started developing in 1999 and which is now open source
free software. The legacy code seems to be in Java. I think with an
agenda that complicated, there was no way to fit SchoolTool in, even if
the programmers are up to getting to grips with your unfinished system
(which I believe they would be).

I have quite a lot of experience with building large enterprise systems
on a shoestring, so I understand the sensitive stages SchoolTool is
currently experiencing. Especially the abstract code is very scary,
because I have seen a lot of abstract code get thrown away at the end of
the day. It's important to keep code in use, by the end user (be that
the developer using the framework or the "actual" end user).

It is quite alarming to find two projects with such similar aims and
such similar technologies, both in the open source arena, but not
cooperating. You have provided the reason - Janastu is obviously about
getting results quickly for specific schools in the Indian context.
Further that project leverages legacy code and concepts which have an
established user base that directly feed into their project.

I am certain a future version of SchoolTool will provide compatibility
one day.

Best regards, Ron

Post by Tom Hoffman

Post by Ronald Wertlen
PS Sorry about any confusion from the last post (bit rushed). More
clarity/info on my blog...
http://ekhayaict.com/wordpress/?p=78

Hi Ronald,
I flat out missed this email, and I still can't find your "last post."
So sorry for the slow response.
I don't think there is much we can do at this point. For better or
worse we are tied to Zope 3, and our codebase is rather abstract and
sophisticated. SchoolTool started with very ambitious objectives to
create an extensible, i18n and L10n-able framework for schools around
the world. This is enterprise software.
The problem is that right now the value proposition for a project like
Janastu to use SchoolTool is at a low point. The codebase is complex,
individual components work and are being tested and used in a handful
of schools, but we are still in the process of pulling this together
into a coherent suite.
So for web developers without heavy object oriented and component
design experience, getting up to speed on SchoolTool is a big step for
an uncertain benefit. It is also an uncertain benefit on our side.
We have a pair of developers. If they're spending time on IRC the
next six months trying to tutor programmers on the fine points of
ZCML, they aren't finishing SchoolTool 1.0 beta (due in October).
There is a long term plan for solving this problem. First off, more
complete releases of SchoolTool make its value more certain.
Secondly, I'm hoping Grok 1.0 http://grok.zope.org/) and SchoolTool
1.0 come out around the same time (i.e., next April), and we'll be
able to start leveraging the ease of development Grok brings to Zope 3
for people looking to extend and modify SchoolTool.
In the meantime, however, our priority is to keep our heads down and
finish SchoolTool.
Having said all that, I'm always happy to offer advice and feedback to
any open source SIS or MIS project.
--Tom

[Ronald Wertlen]

Hi Dinesh,

Thanks for keeping the conversation going. And thanks for correcting my
understanding of your code. As I mentioned in an earlier mail, we are at
some point soon going to have to perform such an evaluation (available
open source school management software) too, so it will be interesting
to see how things have moved along by then. It will probably be toward
the end of the year.

I am sure Tom will answer your organisational question (lead person
leaving), on my part, I find it a bit worrying that they are building
"enterprise software" with just a handful of part-time programmers and
two schools. But I think that Ignas is almost a full-time programmer,
judging by the amount of activity. So I hope the planning works out.

How many programmers do you have working on your project? (Head count/time)

You provide a good example and I am sure SchoolTool is also concerned
about trust. What is your take on internationalisation - I read that you
are translating your software into local languages, how is that going?

Best regards,
Ron

HI Ronald, Tom
Janastu is not tied to any legacy code. We just use our experience of many years
in developing a new code base. We have moved on with Python (Django framework).
I cannot imagine diverting resources at this point in understanding
and/or working
with the state of SchoolTool, as mentioned by Tom. Before starting this new code
base work, we looked into SchoolTool and decided it was not available for our
immediate needs - mostly we were not sure of the development support behind
it (personally, it looked as if the main people behind it has moved on. I might
be wrong and now thanks to Tom, we get a better perspective).
Another objective of the Janastu effort is to provide a tool for
teachers to build
the MIS for their needs. And also to provide a sense of trust for the people
using it. Examples are: a principal can indicate that so and so is an English
teacher and can modify only 8th grade English scores. Also it would be
very hard for even a sys admin to change student scores from the backend.
-d

Post by Ronald Wertlen
Hi Tom,
thanks for the response.
I have looked into the code of Janastu (what is available on the Google
svn) and they are obviously also using a framework. Additionally, they are
tied to a legacy CMS (Pantoto) which the commercial outfit behind the
non-profit started developing in 1999 and which is now open source free
software. The legacy code seems to be in Java. I think with an agenda that
complicated, there was no way to fit SchoolTool in, even if the programmers
are up to getting to grips with your unfinished system (which I believe they
would be).
I have quite a lot of experience with building large enterprise systems on
a shoestring, so I understand the sensitive stages SchoolTool is currently
experiencing. Especially the abstract code is very scary, because I have
seen a lot of abstract code get thrown away at the end of the day. It's
important to keep code in use, by the end user (be that the developer using
the framework or the "actual" end user).
It is quite alarming to find two projects with such similar aims and such
similar technologies, both in the open source arena, but not cooperating.
You have provided the reason - Janastu is obviously about getting results
quickly for specific schools in the Indian context. Further that project
leverages legacy code and concepts which have an established user base that
directly feed into their project.
I am certain a future version of SchoolTool will provide compatibility one
day.
Best regards, Ron

Post by Tom Hoffman

Post by Ronald Wertlen
PS Sorry about any confusion from the last post (bit rushed). More
clarity/info on my blog...
http://ekhayaict.com/wordpress/?p=78

Hi Ronald,
I flat out missed this email, and I still can't find your "last post."
So sorry for the slow response.
I don't think there is much we can do at this point. For better or
worse we are tied to Zope 3, and our codebase is rather abstract and
sophisticated. SchoolTool started with very ambitious objectives to
create an extensible, i18n and L10n-able framework for schools around
the world. This is enterprise software.
The problem is that right now the value proposition for a project like
Janastu to use SchoolTool is at a low point. The codebase is complex,
individual components work and are being tested and used in a handful
of schools, but we are still in the process of pulling this together
into a coherent suite.
So for web developers without heavy object oriented and component
design experience, getting up to speed on SchoolTool is a big step for
an uncertain benefit. It is also an uncertain benefit on our side.
We have a pair of developers. If they're spending time on IRC the
next six months trying to tutor programmers on the fine points of
ZCML, they aren't finishing SchoolTool 1.0 beta (due in October).
There is a long term plan for solving this problem. First off, more
complete releases of SchoolTool make its value more certain.
Secondly, I'm hoping Grok 1.0 http://grok.zope.org/) and SchoolTool
1.0 come out around the same time (i.e., next April), and we'll be
able to start leveraging the ease of development Grok brings to Zope 3
for people looking to extend and modify SchoolTool.
In the meantime, however, our priority is to keep our heads down and
finish SchoolTool.
Having said all that, I'm always happy to offer advice and feedback to
any open source SIS or MIS project.
--Tom

[Tom Hoffman]

Post by Ronald Wertlen
I am sure Tom will answer your organisational question (lead person
leaving), on my part, I find it a bit worrying that they are building
"enterprise software" with just a handful of part-time programmers and two
schools. But I think that Ignas is almost a full-time programmer, judging by
the amount of activity. So I hope the planning works out.

This year we have two full time programmers, Ignas and Alan Elkner.

--Tom