Mailinglist Archive: yast-devel (177 mails)

< Previous Next >
Re: [yast-devel] Direction of YaST Architecture?
Am Thu 10 Feb 2011 03:47:56 PM CET schrieb Klaus Kaempf <kkaempf@xxxxxxx>:

* Jiri Suchomel <jsuchome@xxxxxxx> [Feb 10. 2011 15:37]:

But it is there now. And the result is, webYaST brings huge amount of packages
because of YaST.

Yes :-( Thats why we have to move away from static languages like YCP
to dynamic ones like Python or Ruby. Then a failing 'import' can be
handled gracefully and package dependencies are drastically reduced.

Using 'dlopen' and 'dlsym' as the killer feature of dynamic languages is really not fair. Please be honest in promoting Ruby.

There is IMO only 1 argument supporting usage of dynamic languages:
Program can be inspected and changed while running in production.

And this ^^^ is only useful for always running services. Erlang excels in this. But it is really not the use case of YaST.

Garbage collection, simple syntax, working with collections or ability to express ideas clearly is not just domain of Duck Typed languages.

OTOH good luck with establishing a stable interface in a Duck Typed language. This can only be done with lots of documentation, convention and tests. But it cannot be enforced or verified.
Python solved interfaces with zope-interface, which allows the programmer to specify "how do the data look like". Duck Typing by itself doesn't need to be bad. Going all the way and being unable to describe data structures definitely is. Most importantly because data structure description is an important communication channel for programmers.

Martin


Klaus
---
SUSE LINUX Products GmbH, GF: Markus Rex, HRB 16746 (AG N├╝rnberg)

--
To unsubscribe, e-mail: yast-devel+unsubscribe@xxxxxxxxxxxx
For additional commands, e-mail: yast-devel+help@xxxxxxxxxxxx




--
To unsubscribe, e-mail: yast-devel+unsubscribe@xxxxxxxxxxxx
For additional commands, e-mail: yast-devel+help@xxxxxxxxxxxx

< Previous Next >
List Navigation
This Thread