search

zooniverse / Lens-Zoo

Apache License 2.0
0 stars 4 forks source link

Candidate lists on Discoveries pages #265

Closed drphilmarshall closed 9 years ago

drphilmarshall commented 9 years ago

Rough suggested plan:

anupreeta27 commented 9 years ago

@drphilmarshall - is this the page where we plan to put ALL (new+known) candidates with G >0 not just G>2. If yes, I already have made small cutouts centered on the candidates (.pngs). I just need to know, if I should make the tar available somewhere and to who -or- should I convert it into a gallery like in paper 2 (an array of annotated pngs split over several pdf pages). In any case, pngs should be annotated with similar info as we have in the paper.

drphilmarshall commented 9 years ago

To answer this properly we need to decide who our target audience is, and relatedly what we want to host on the team website (currently at http://drphilmarshall.github.io/SpaceWarps/). My feeling is that the app should mainly function as advertising, and mainly to the volunteer community - in this model it would show images of new candidates, hopefully linked to Talk, present brief summaries of projects and links to papers, and then provide links that people can follow to find out more - and the place where we would put this extra material would be the team website (where we can change things around faster, and the editing is simpler!).

Whatever we decide, I agree that having annotated small cutouts on the app results page would be good - mainly because these particular images will get copied and passed around the internet, so having them display their own provenance would be good. Some of this can be achieved with a very informative filename (eg CFHTLSJ020642-095157_SpaceWarps_gri_10x10arcsec.png). For the annotation itself, I'd go with the object name (CFHTLSJ020642-095157) at the top, and http://spacewarps.org along the bottom, and drop the SW ID and grade. We probably want to use smaller font and gray text, to make the annotation more subtle (so that people want to use the images!). What do you think?

From a practical point of view, using our own cutouts means either uploading PNGs to the Lens-Zoo site, or pulling them in from the team website. I guess if we made a compete gallery on our own website, the easiest thing to do would be the latter (just write links into the app results page).

On Wed, Apr 22, 2015 at 1:07 AM, anupreeta27 notifications@github.com wrote:

@drphilmarshall https://github.com/drphilmarshall - is this the page where we plan to put ALL (new+known) candidates with G >0 not just G>2. If yes, I already have made small cutouts centered on the candidates (.pngs). I just need to know, if I should make the tar available somewhere and to who -or- should I convert it into a gallery like in paper 2 (an array of annotated pngs split over several pdf pages). In any case, pngs should be annotated with similar info as we have in the paper.

— Reply to this email directly or view it on GitHub https://github.com/zooniverse/Lens-Zoo/issues/265#issuecomment-94978367.

drphilmarshall commented 9 years ago

Fastest possible route to images on the page = link to paper 2 gallery png.

drphilmarshall commented 9 years ago

Hey @chrissnyder thanks for fixing up the formatting! The hyperlinked VICS82 candidate list looks pretty good I think :-) We'll work on getting a CFHTLS gallery up in the same style. Cheers!

anupreeta27 commented 9 years ago

@drphilmarshall see this new folder with annotated pngs of all SW candidates https://github.com/drphilmarshall/SpaceWarps/tree/master/analysis/projects/CFHTLS/candidates

drphilmarshall commented 9 years ago

Thanks @anupreeta27!

Here are some suggestions for formatting in a second iteration - can you please:

Thanks!

On Mon, Apr 27, 2015 at 6:12 AM, anupreeta27 notifications@github.com wrote:

@drphilmarshall https://github.com/drphilmarshall see this new folder with annotated pngs of all SW candidates

https://github.com/drphilmarshall/SpaceWarps/tree/master/analysis/projects/CFHTLS/candidates

— Reply to this email directly or view it on GitHub https://github.com/zooniverse/Lens-Zoo/issues/265#issuecomment-96508655.

anupreeta27 commented 9 years ago

@drphilmarshall thinking about this a bit more after your suggestions, i've made some changes - the new annotation looks like this - let me know if you are fine with this.

cfhtlsj020122-091515_spacewarps_asw00010he_1 3_30asec_gri

anupreeta27 commented 9 years ago

ps: i will switch the names to this format - CFHTLS JHHMMSS.s+DDMMS

drphilmarshall commented 9 years ago

I like the subtlety (and the font! :-)

Didn't we decide to include the grade on the image? Or do you think it's enough to propagate it in the filename? (BTW "grade0.3" might be better than just "0.3" there)

If we leave it off, I'd vote for name on the left, URL on the right (names should go first, I think). With the longer name, you might have to go back to top and bottom lefthand corners for the two annotations.

On Tue, Apr 28, 2015 at 4:44 AM, anupreeta27 notifications@github.com wrote:

ps: i will switch the names to this format - CFHTLS JHHMMSS.s+DDMMS

— Reply to this email directly or view it on GitHub https://github.com/zooniverse/Lens-Zoo/issues/265#issuecomment-96890073.

anupreeta27 commented 9 years ago

i left the grade out since you said these are mostly for use on the web. the filenames will still have the grade, in case people want to use them for sorting the sample.

anupreeta27 commented 9 years ago

@drphilmarshall this list says http://cds.u-strasbg.fr/vizier/Dic/formats.htx that the format - HHMMSS+DDMMSS - this is what i had earlier, why not stick to this?

anupreeta27 commented 9 years ago

i mean "CFHTLS HHMMSS+DDMMSS" is an acceptable format - why not stick with this?

anupreeta27 commented 9 years ago

@drphilmarshall you can find the updated "candidates" dir under analysis/projects/CFHTLS

drphilmarshall commented 9 years ago

OK, submitting a pull request (to see if @chrissnyder can help with one last bit of eco fixit magic). Thanks for the images, Anu!