Daniel Mietchen's dashboard

Daniel Mietchen replied on June 26, 2009 01:48 to the idea "User-defined source publications" in Pubfeed:

They are different, and so both potentially useful for Pubfeed. eTBLAST would be better for the initial seeding and cumbersome for further trimming, while PubMed could go, like Google Scholar or DBLP, for both the initial seeding and the repetitive refinement.

But if you ask for priorities, I would go first for the above-mentioned option of refining a seed corpus with further manual searches. Second would be the option to start from eTBLast, third would be including PubMed (or HubMed).

I would appreciate if you would share some usage stats with me (you have my email), since I have started bringing in some more people here, and plan to intensify this.

Daniel Mietchen replied on June 25, 2009 15:40 to the idea "User-defined source publications" in Pubfeed:

Another way to reduce the load on Google Scholar while increasing the sensitivity would be to allow eTBLAST results to be imported into pubfeed - these can very easily be tailored. Example at http://invention.swmed.edu/etblast/us... .

Daniel Mietchen replied on June 24, 2009 09:12 to the problem ""Populating feed" takes ages" in Pubfeed:

It's a pity but very understandable that they limit you in query frequency. Anyway, if there is the previously discussed option to do a sequence of supplementary manual searches and to populate the seed list with some of the items found this way, the lists can be constructed much more efficiently. Currently, I am keeping the seed list as large as possible in the hope that some of the papers I am waiting for will eventually pop up. By the way, I would appreciate some way of direct messaging, perhaps via @EvoMRI (I am following you already).

Daniel Mietchen replied on June 22, 2009 19:04 to the problem ""Populating feed" takes ages" in Pubfeed:

Same problem again with http://pubfeed.cs.toronto.edu/f/156/m... .

Daniel Mietchen replied on June 20, 2009 03:16 to the question "Shouldn't their be a link to http://microrevie.ws/ somewhere on this page?" in Notches:

Well, the feedback button that led me here regularly appears when I click on a non-link item in their microreviews and then the notch icon. And should I continue to post suggestions on @hreviews here or elsewhere? I would particularly be interested in feedback on scientific uses, as discussed at http://bit.ly/3WgwOM .

Daniel Mietchen replied on June 20, 2009 03:11 to the idea "Expand shortened URLs from tweet to review" in Notches:

Sounds good, Tim! And while we are at it, a little validity check of the expanded link may help the review - perhaps by sending an automated reply to the tweeter that the link was wrong. Thanks!

Daniel Mietchen asked a question in Notches on June 19, 2009 21:45:

Shouldn't their be a link to http://microrevie.ws/ somewhere on this page?

Daniel Mietchen shared an idea in Notches on June 19, 2009 18:53:

Expand shortened URLs from tweet to review

Example here.

This would allow to aggregate reviews targeted at the same URI.

A comment on the idea "User-defined source publications" in Pubfeed:

I repeat support for my comment above - this would greatly increase usability. After that, an option to hide some of the hits (especially ones seen during previous "populations" of a feed) would help to trim down the list to a useful corpus. – Daniel Mietchen, on June 19, 2009 16:56

Daniel Mietchen asked a question in Pubfeed on June 19, 2009 14:43:

Linking from welcome page

Shouldn't http://getsatisfaction.com/pubfeed have a link to http://pubfeed.cs.toronto.edu/ ?

Daniel Mietchen replied on June 19, 2009 14:41 to the problem "RSS" in Pubfeed:

Thanks. Having links to the search results would come in handy and allow to make use of the "include entry description as comment" option below.

Daniel Mietchen replied on June 19, 2009 14:12 to the problem "RSS" in Pubfeed:

I made some appear, but they had the wrong outgoing link. Example at http://ff.im/4b4zP .

Daniel Mietchen replied on June 19, 2009 10:14 to the problem "RSS" in Pubfeed:

XML template still contains "example.com". Example: http://bit.ly/kBbNd - the "auditory-motor coupling" link in "...ago from Auditory-motor coupling" leads to http://example.com/f/132/auditory-mot... , even though the feed comes from http://pubfeed.cs.toronto.edu/f/132/a... .

Daniel Mietchen marked one of Andrew Trusty's replies in Pubfeed as useful. Andrew Trusty replied to the problem "Starts via Google Scholar _very_ slow".

A comment on the idea "User-defined source publications" in Pubfeed:

Another way to go about this would be to have the possibility, after creating a feed as usual, to do independent searches, from which the results can be added to the feed. This has the advantage that only valid search results will be used to seed the feed. – Daniel Mietchen, on June 05, 2009 20:49

A comment on the problem "RSS" in Pubfeed:

OK; it does work now. Thanks! – Daniel Mietchen, on June 05, 2009 10:43

A comment on the problem "Starts via Google Scholar _very_ slow" in Pubfeed:

They are usually quick in their decisions and supportive of endeavours like this one. For an example of a site which got such permission, see http://toolserver.org/~verisimilus/Sc... . – Daniel Mietchen, on June 05, 2009 10:37

A comment on the problem "Starts via Google Scholar _very_ slow" in Pubfeed:

This note refered to the situation that I could not even get past the "Start by telling me what your favorite papers are." stage, since none of the searches I tried did get populated, and the search indicator continued to rotate for about ten min (I then posted this problem and closed the widows).

Seems to run smoothly now. In other feeds, I have used up to 140 source publications, with no apparent problems related to that. Still experimenting, though. – Daniel Mietchen, on June 05, 2009 09:14

Daniel Mietchen reported a problem in Pubfeed on June 05, 2009 03:26:

Starts via Google Scholar _very_ slow

Haven't succeeded in starting any new feed this way, though I noticed that the existing ones have been much improved after GS was made available.

Daniel Mietchen replied on June 05, 2009 02:40 to the idea "User-defined source publications" in Pubfeed:

Thanks. I think that Google Scholar is a better place to start from for most subjects.