Hi LimeWire development/suport team,

Nifty neat program , free of ads, well done.

I could just use this and not give any feedback, but I see it has mentioned a few times on the website that you would like feedback-- >

A few problems and suggestions I have to talk about:
I have downloaded a file exmaple "oakenfold.mp3"
I have about 30 megs of the file....
Now i log on again the next day and it cant resume the file , it cant find sources.

This is what irritates me , because how come I can find sources myself. I say this because when i start a new download of the same mp3 , even though it is a different file name or description, it starts downloading..... and if i listen to the mp3 it is the same as the one i have (the 30 meg one) -- >

My issue here is , why do I have to download the whole mp3 again? I have 30 megs , so i dont want to download it from scratch again, it is a waiste of time.

I have tried a few tricks with renaming files and copying them and stuf but it doesnt work and I also received a message saying that the file is not a valid incomplete file.


I want to suggest that something similar to the download technique that Napster used or even Getright or Netpumper...

unless there is a way to use the file i have(30 megs) and then use that to resume with? but how.... like i mentioned above , i have tried but i have no success.

Please assist or let me know what you think about what i have mentioned.

Kind Regards
MusicToad

Hi MusicToad. I'm not with LW, but perhaps this info will help:

Sounds like you are describing a case where two files are similar, but not identical.

Here's the long-winded explanation. Each file shared on the gnutella network will have a unique identifier calculated ("hash"), so that if even 1 bit changes, the file will be treated as a different file. This is GOOD--since it allows pieces of the same file to be merged from multiple hosts ("swarming") and prevents the malicious corruption of good files. True, most of us wouldn't notice if just a few bits of a picture changed, but it could make a program file unusable.

So if a download of a file with one hash starts and the host goes offline, a search might find other files with the same/similar names, but a differrent hash. The client (LW in this case) knows it is not exactly the same file, and so starts a new download.

btw--you can check this by clicking on the file in the library's incomplete folder, and right-clicking to call up either the bitzi or magnet details to see the "sha-1 hash."

I see what you are saying, so it sounds like this simple terms:
the application probably writes a piece of info to the header or footer of the file, and thats how it knows what is what. I reckon they need to make some type of a resume feature and allow the user to accept or not.

I mean especially for phone line users man, i mean it is a pain having to download 30 megs again, or even 20 , i mean some of the people stilll pay rates for being connected, whether it be per minute or 10 minutes or half hour or 1 hour. The rates dont appply for everyone as they do in the STATES.

Anyways, *sigh* i need to figure out what i can do for resuming, i wish I could........and this is until the LIMEWIRE group figure out a work-around.

I am sure I have a valid point here for those that would also like files resumed.....

I dont know ? am i wrong?

Anyways, thanks alot for that explination.

Regards