[nzlug] (slightly)OT - huge OSS opportunity

[David McNab]

anru chen wrote:
> we have a local php developers mail list, probably a better place to
> rant about business software, and many people in that list are
> business owners. (phpug at phpug.org.nz).

Doesn't feel quite right, since by its name one would expect it to be
very pro-php. The list owners might not appreciate people slagging PHP.

> from my point of view, a winning groupware does  not need that ( OO
> RAD web app framework like Ruby on Rails, Python's TurboGears or
> similar good Java  equivalents) to be a winning groupware, what we
> need is a good solution.

>From my point of view, in the long run, we need both.

> all of those new tech like AJAX,ROR, TurboGears, they are just lastest
> buzz words.
> what kind of new thing they have bring to developer ?

AJAX brings more to the user than to the developer - the ability to talk
to a web host, get a response and update part of the page without having
to refresh the entire page. Not functionally essential, but a major
boost for the comfort of the user.

ROR and TurboGears are RAD Web App frameworks, that let people get apps
up much more quickly than with raw php, and (if working under time
pressure) end up with much clearer and more maintainable code.

> just some old
> techs changed to a new name.

That's a bit sweeping IMHO

> if you need a customized groupware, better find some good developers.

I'm working on some code to talk xmlrpc to egroupware, with the
possibility of writing a nice wx client to deliver much quicker
performance, and fill in for egroupware's shortcomings. If/when I get
that fully happening, then sugarCRM is headed for the bit bucket.

Cheers
David

> 
> 
> regards,
> 
> anru
> 
> On 8/26/06, David McNab <david at rebirthing.co.nz> wrote:
>> Hi,
>>
>> (apologies for OT post - this is about OSS and business, not
>> specifically linux)
>>
>> Over the last few days, I've been going through dozens of web-based
>> groupware apps looking for something to suit our consultancy business.
>>
>> <rant>
>>
>> The bottom line - all the web-based groupware apps suck!
>> SugarCRM sucks slightly less, so I've gone with that. But it still has
>> major problems.
>>
>> The groupware development scene is a fascinating phenomenon. Sourceforge
>> and Freshmeat list hundreds of groupware apps, which vary only in their
>> position on the suck spectrum, various places ranging from 'severely' to
>> 'extremely' to 'impossibly'.
>>
>> It seems that people start up a new groupware app, aiming to come up
>> with something less painful than the existing offerings, but end up with
>> something just as bad or worse.
>>
>> So what I'm saying is that IMHO there exists a tremendous opportunity
>> for someone with great web app dev skills to steal a huge chunk of the
>> market.
>>
>> Firstly, I need to get specific about the existing groupware suckage:
>>
>>  * all groupware apps do one or two things well, maybe many things well,
>>    but no app does everything well
>>
>>  * the set of features I'm looking for is diffused among many
>>    different groupware apps, and not available in a single app, features
>>    such as:
>>     - composite to-do tasks (tasks with dependencies)
>>     - ease of sharing tasks, appointments, contacts etc between users
>>     - good level of AJAX usage
>>     - categories management - allow categories to be assigned to:
>>         - contacts
>>         - appointments
>>         - to-do items
>>     - bleeding obvious - a 'schedule meeting' button on the display of
>>       a contact's record, which sets up a meeting with that contact,
>>       and avoids the name to type the contact's name into the 'subject'
>>       field for the meeting
>>     - rich-text editing for emails, notes etc
>>     - bulk-mailing (not spamming) to opt-in clients
>>     - good xml-rpc or SOAP interface
>>     - product catalogue, with order processing
>>
>>  * nearly all the existing OSS groupware apps are written in PHP, which
>>    is one excruciating language which by its very structure tempts
>>    people into poor design/programming practice, introducts exponential
>>    growth of complexity and hikes up development costs.
>>
>>    A winning OSS groupware app should use an OO RAD web app framework
>>    like Ruby on Rails, Python's TurboGears or similar good Java
>>    equivalents
>>
>>    Also, a good groupware app should have facilities for users to set
>>    up custom views, forms etc
>>
>> </rant>