A few Limewire sessions will teach you a depressing lesson about permutations: a popular mp3 or vid will bear about as many names as host computers -- with or without hypens, dashes, parentheses, capitalization and bad spelling. While Limewire can group these files as identical, once the download begins, you are locked into that name. You download half of a larger vid under one name, can't find that name after the host goes offline, so you download another, or the "smart" download simply connects to another host, and you see the 50% drop to 0%. I think the highest number of names I had for one video was seven. This makes for a lot of useless clutter in the library.

Now if Limewire is smart enough to recognize these files as identical on the front end, why not with resumed downloads? Just a little button saying "Same file" would do. Since I assume downloads are resumed by finding the last completed hex code of a file, why couldn't it do that for files identical except for variations on the names if you tell Limewire to do so?

Excellent idea - which is why it is being implemented as we type.

Hashing will allow LimeWire to tell whether two files are identical irrespective of name. Hopefully the implementation will drastically improve downloads - particularly of larger files.

However, this isn't a trivial modification (it is rather more sophisticated than your description) so it may not come out with the next version - although I'm hoping it will.

Mark

I am utterly clueless when it comes to how downloads are resumed; the hex thing was just a guess. Can you give me a URL for somewhere where I could find out about it?

The best I can suggest is starting from the LimeWire homepage and following the links for developer resources or open source.

Mark