Oreilly CodeZoo

R

richard

gene said:
http://radar.oreilly.com/archives/2005/08/codezoo_program_1.html

http://python.codezoo.com/

Nice to see that python is in-demand, but what is the rationale for
another ASPN cookbook/Parnassus / pypackage / dmoz type repository?

We (PyPI / Cheese Shop developers) are talking to the CodeZoo people about
the relationship between the two systems. Things PyPI has:

1. python setup.py register
2. python setup.py (sdist|bdist|bdist_egg|bdist_wininst|...) upload
3. http://cheeseshop.python.org/
4. better categorisation (IMO)
5. XML-RPC interface
6. 852 packages registered

The things that CodeZoo has:

1. reviews, user tips and ratings
2. a full-time editor to make sure that the reviews are reasonable
3. links to O'Reilly articles
4. a big DOWNLOAD button. hurm, that's a good idea, I should add one of
those to the PyPI interface... I hope no-one has that patented
5. DOAP (being added to PyPI)
6. Ads in the top bar and a sidebar (hehe)

Getting links from PyPI over to the CodeZoo information would be pretty
easy. Getting links from the CodeZoo pages to PyPI information / packages
would be pretty easy.


Richard
 
G

gene tani

There it is, right on homepage. Thanks, it looks quite nice, and i see
you'll be an indep. gems server, and more... I did try searching on
"postscript" and it didn't pull up pyscript. Also, HTML on the
"Search" results page is kind of not rendered great in Firefox1.06
/WinXP SP2. BUt it's a really great start overall.

http://python.codezoo.com/news.csp
 
J

Jorge Godoy

richard said:
We (PyPI / Cheese Shop developers) are talking to the CodeZoo people about
the relationship between the two systems. Things PyPI has:

1. python setup.py register
2. python setup.py (sdist|bdist|bdist_egg|bdist_wininst|...) upload
3. http://cheeseshop.python.org/
4. better categorisation (IMO)
5. XML-RPC interface
6. 852 packages registered

Richard,


Some feedback:


With regards to 6, it would be interesting, though, to separate the
classification using just the software name and not the version. For
example http://cheeseshop.python.org/pypi?:action=browse&asdf=256 has four
entries to ZODB3, in four different versions. This would decrease the
amount of packages registered, but would allow a more realistic view of
"packages" and not of "package versions", IMVHO.

A subtree with version numbers could solve the problem of storing
information for old versions when a new one is added.

With regards to 4, I'd like to see the "Topic" category in the beginning of
the page -- this is what most people search, I guess -- and not at the
bottom of it. I'm usually looking for a software to do <something> and not
a software that is being planned, is in alpha - beta - production stage,
etc.

Also, there's a discrepancy on the amount of packages stored. The
categorization page says there are 679 packages. The main page says there
are 852. I believe the categorization page does what I suggested and
disregard different versions of the same software. I think this (679) is
the correct number.

Ah! And an RSS feeder would be interesting, to know when there are new
packages or when a package has been upgraded... ;-) Where's the
"wishlist"? :)

I hope you see this as a constructive feedback from someone that has just
screened the pages, getting to know it.


Be seeing you,
 
R

richard

Jorge said:
A subtree with version numbers could solve the problem of storing
information for old versions when a new one is added.

With regards to 4, I'd like to see the "Topic" category in the beginning
of the page -- this is what most people search, I guess -- and not at the
bottom of it.

That's a good idea. Unfortunately, changing the sorting of the classifiers
will be fun. Could I ask you to submit an RFE "bug" via the link on the
pypi page?

Also, there's a discrepancy on the amount of packages stored. The
categorization page says there are 679 packages. The main page says there
are 852. I believe the categorization page does what I suggested and
disregard different versions of the same software. I think this (679) is
the correct number.

Yep, this is related to your other comment regarding my 6th point. Also, I
don't believe packages are included in the browse if they have no
classifiers.

Ah! And an RSS feeder would be interesting, to know when there are new
packages or when a package has been upgraded... ;-)

There's a top-level RSS feed, but not one per-package.

I hope you see this as a constructive feedback from someone that has just
screened the pages, getting to know it.

Thanks!


Richard
 
J

Jorge Godoy

richard said:
That's a good idea. Unfortunately, changing the sorting of the classifiers
will be fun. Could I ask you to submit an RFE "bug" via the link on the
pypi page?

I'll do that tomorrow morning.
Yep, this is related to your other comment regarding my 6th point. Also, I
don't believe packages are included in the browse if they have no
classifiers.

It means that some packages might be "lost" somewhere? How about an
"unclassified" classifier? :)
There's a top-level RSS feed, but not one per-package.

I was thinking about one per topic / classification, not per package. It
would be overkill to subscribe to 600+ feeds.


Be seeing you,
 
R

richard

Jorge said:
I'll do that tomorrow morning.

Thanks!

It means that some packages might be "lost" somewhere? How about an
"unclassified" classifier? :)

No, it means that when you're browsing by classifier, the packages that
don't specify any classifiers won't match your browse specifications ;)

I was thinking about one per topic / classification, not per package. It
would be overkill to subscribe to 600+ feeds.

Hurm. Interesting idea. Could be done with a bit of work.


Richard
 

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments. After that, you can post your question and our members will help you out.

Ask a Question

Members online

Forum statistics

Threads
473,744
Messages
2,569,483
Members
44,901
Latest member
Noble71S45

Latest Threads

Top