[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Liberty-eiffel] Marketing Liberty-Eiffel

From: Raphael Mack
Subject: Re: [Liberty-eiffel] Marketing Liberty-Eiffel
Date: Sun, 15 Sep 2013 22:21:57 +0200


Hans, thanks for sharing your ideas. I comment on a few of them.

Am Sonntag, den 15.09.2013, 19:08 +0200 schrieb H. Zwakenberg | Ocean
Consulting GmbH: 
>  - even if only a few of the old guard contemplate to 'perhaps' come 
>  back, it might be a good idea learn in advance what made _them_ stop 
>  using SE and under what conditions they would return to SE's successor.  
>  It's irrelevant what _we_ think their reasons were, we need to learn how 
>  _they_ feel about it.

More interesting than the core dev team would be the user base in my
eyes. Here I know from a few that the reasons are:
- the business decided to leave eiffel in favor of python
- new compiler version broke too much code
- the core dev team was too arrogant

>  - I seem to recollect that back in the SE-days, someone took the effort 
>  to start implementing the Language Shootout modules.  If LE is as good 
>  as we hope it to be, this would be a platform to show it off.  Is the 
>  old SE code still around and could it perhaps be a basis for improvement 
>  (and killer shootout scoring)?

LibertyEiffel should show up there, yes. I guess it's on Paolos ToDo

>  - We should start to collect email addresses of IT magazine editors, to 
>  be able to pinpoint press-releases.

I don't expect too much of this, but maybe there is a chance, that a it
magazine prints an Eiffel introduction, based on a tutorial we
provide... Anyhow, we should improve to become a real
tutorial to learn the language from scratch.

>  - Perhaps we should create a separate distribution for educational 
>  institutions.  Such a distributions should contain material that those 
>  that lecture can use right away

Here I guess it's too much work for us with the expectation, that it
will not be used anyhow ;-(

>  - For LE to be generally usable, I reckon we don't just need the 
>  compiler, but a GUI and database library as well. Ideally, we'd have a 
>  generic layer for each, that allows different GUI's and different dB's 
>  to be used as plugins.

This is why the wrappers are important. To be able to use many existing 
libraries for GUI etc. Providing an own liberty library seems impossible due to 
lack of manpower. The same applies for the IDE. Cyril and I use emacs as 
editor, but I think in the long run we should try to get an eclipse integration.


reply via email to

[Prev in Thread] Current Thread [Next in Thread]