Brief article on benchmarks of Python repository with leading DVCSen

[Alexander Belchenko]

David Cournapeau пишет:
> Alexander Belchenko wrote:
>> Nicholas Allen пишет:
>>   
>>>> I don't know if the steps in parentheses are still needed, but they
>>>> certainly were, if you wanted to avoid extremely slow initial clone
>>>> times, at one point (and I haven't seen anything saying that's changed
>>>> - nope, I went and checked, it's still true that this is the
>>>> documented approach).
>>>>   
>>>>       
>>> You could do a lightweight checkout that would not need all the history
>>> or you could create a stacked branch. Both of these should significantly
>>> reduce the bandwidth and download time. If the smart server was running
>>> on the Python site I believe it basically does the whole tarball thing
>>> for you without any change in commands you execute.
>>>
>>> From a benchmarking point of view it would also make sense to benchmark
>>> smart server against smart server. If Bazaar is just using a dumb file
>>> storage then it does not make sense to compare the performance of that
>>> to git, say, which can only use a smart server AFAIK.
>>>     
>> According to manpages git can use dumb http server, but in practice I think it's very rare.
>>   
> 
> It is not that rare: it is very useful for people behind proxies (like
> me :) ). It is very slow, though, almost as slow as bzr.

I guess you actually mean: bzr is anyway slower.
But IIUC to use git in dumb mode you have to update special files 
regularly in your repo on the server to allow dumb mode to work 
properly, is not?