User talk:NeilFraser

Brain not getting enough excercise (sic) (or sick)
Wow, I can't believe how many misspellings of exercise you're finding in assorted dog-related articles! Since I try to vet them all for basic grammar and spelling, I'm mortified to see this crop up over and over in articles I've proofread. I think I'll have to turn in my proofreading secret decoder ring and badge. Thanks for the improvement and making us look more literate overall. Elf | Talk 04:26, 2 February 2006 (UTC)


 * No problem. Glad someone appreciates the Spelling Nazi in me.  :)  NeilFraser 06:02, 3 February 2006 (UTC)

Dental lamina
Thank you for fixing my typos. That is horrendous I wrote all those mistakes in such a small article. I had to revert two of your edits, and I wanted to point them out in case you wonder why I did that. The correct term in this case is not "reabsorbed" but "resorbed". The term is used in reference to biologic tissues. You can check here if you want for a better dictionary explanation. Thanks again for fixing all my mistakes. - Dozenist talk  15:01, 4 February 2006 (UTC)


 * Thanks for the note! 'Resorb', one learns something every day. NeilFraser 03:13, 5 February 2006 (UTC)

Diff
Hi Neil,

I have had a look at your diff algorithm and the cited paper. If I understand it correctly, your algorithm is not able to detect block moves, be it switched words or whole moved paragraphs. That is an easy task for Heckel's algorithm (the name is misspelled on http://neil.fraser.name/news/2006/jsdiffwiki.html) on which my implementation is based (see User:Cacycle/diff). If a block has been moved, your implementation would detect an insertion and a deletion while in fact not a single word has been changed. I would also think that in general Heckel's algorithm is much faster than your algorithm.

Of course it is very easy to design a synthetic text where Heckel's algorithm does not work: simply take a text without unique words, like in the example on your page. Fortunately, this happens extremely rarely in texts of natural language. As Heckel wrote in his paper: there is not a general diff algorithm, simply because there is no single definition of a text difference.

Cacycle 22:46, 23 March 2006 (UTC)

Help Request: Lost And Found (Christian Rock Band)
I've been trying to edit the photographs so they'd be better and not in a long line, but everything I do just causes them to overlap and makes the page look like a mess. Can somebody please help? http://en.wikipedia.org/wiki/Lost_and_Found_%28Christian_rock_band%29 Stein Auf! KagomeShuko 05:37, 28 March 2007 (UTC)

Taskforce
Psdubow 14:53, 10 July 2007 (UTC)

ArbCom elections are now open!
MediaWiki message delivery (talk) 13:40, 23 November 2015 (UTC)