Re: [Monotone-devel] partial pull #2 - gaps instead of a single horizon

[Rob Schoening]

On 6/1/07, Markus Schiltknecht <address@hidden > wrote:

Basically, what I'm stating is, that the avg. history vs. checkout size
ratio probably is that low, because the tools to track history were
lacking. I bet that this ration will grow, as soon as people learn about
the benefits of properly tracking history.

It is true that the ratio of repository size to checkout size will grow (unbounded). 

 

But for the majority of the users for which this is a primary pain point (what the Pidgin wiki calls the "casual developer" who is looking to make a one-off patch to the source code), the size of the _useful_ or _relevant_ portion of the repository to checkout size is likely to be somewhat more constant. 
 

From the Pidgin wiki:

<<Complete (initial) pulls of Gaim are ~35 minutes on a fast connection. This is too long to expect any casual developer to invest. >>

 

Perhaps the kinds of proposals for partial pull that have been discussed are, indeed, the best solutions.   But I do think that you could solve this "casual developer problem" with one of many far simpler solutions.

 

For instance, what if you simply deferred the full sync until *after* the user got his workspace?  That is, don't worry about trying to solve the problem of syncing repositories that look like swiss cheese.  But do allow a user to get the first slice without copying the whole block....

 

RS