Perl 6 - the future is here, just unevenly distributed

IRC log for #opentreeoflife, 2015-05-27

| Channels | #opentreeoflife index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:23 kcranstn joined #opentreeoflife
00:27 jimallman joined #opentreeoflife
09:47 josephwb joined #opentreeoflife
12:49 jar286 joined #opentreeoflife
13:52 kcranstn joined #opentreeoflife
15:58 kcranstn jimallman?
15:58 jimallman kcranstn: hi!
15:58 kcranstn wondered if now is an ok time to chat about collections
15:59 jimallman i’m doing a quick dryad test… 5 min?
15:59 kcranstn sure
15:59 jimallman thanks, i’ll ping you here
16:04 kcranstn might postpone - someone just dropped by
16:05 jimallman ok by me. i’ll be around, just ping me again here
16:16 kcranstn ok, back
16:16 kcranstn thanks for your work on this
16:19 kcranstn I am not convinced about the need for three status options (include / exclude / remember) for trees in collections
16:20 kcranstn seems simpler to have include / exclude as a green / red slider without the popup menu
16:20 kcranstn and then be able to order the table by headings
16:20 kcranstn which allows us to ditch the Show buttons at the top of the list
16:21 jimallman …reviewing this now...
16:22 jimallman ah yes, i see what you mean about the Show All/Included toggle
16:22 jimallman i thought people might want to make decisions slowly, but it’s certainly simpler if they make the call in a single sitting.
16:22 jimallman the decision to include a single tree, i mean
16:23 kcranstn it seems overly complicated
16:24 kcranstn i.e. it seems like us deciding what users might want instead of giving basic features and waiting for feedback
16:24 jimallman my concern was that someone might find an appealing tree, but need to follow up (with its author, i guess) to make the final decision. i suppose that’s overly paranoid, since (in principle) we have the published paper and lots of data to answer these questions.
16:24 kcranstn undecided = exclude
16:24 kcranstn but default should be include
16:25 kcranstn we also need help text someone to explain this stuff
16:25 kcranstn somewhere
16:25 jimallman gotcha. my other concern was that they might keep re-discovering the same tree and ultimately reject it, not realizing they’ve done this already (perhaps more than once). that’s why i was going to retain “excluded” decisions.
16:26 jimallman much simpler if we ignore this possibility. then it’s just a list of trees, not a list of decisions-about-trees.
16:26 kcranstn wait, I am not advocating removing the include / exclude
16:26 kcranstn only the third option (Remember)
16:26 jimallman i see.
16:27 jimallman that still leaves us with the Show All/Included toggle. is that OK for you?
16:27 kcranstn Why? if you can sort the list based on status, this give the same functionality
16:27 kcranstn and takes up less screen real estate on a popup
16:28 jimallman so we’d just always show a list with Excluded trees, perhaps at the bottom? i want to retain a “natural” sort that reflects the ranking of trees in the collection.
16:28 kcranstn Rank column?
16:28 kcranstn the rank should be explicit, not dependent on sort order
16:28 jimallman yes, i show this now (ghosted numbers at left).
16:29 kcranstn probably want move buttons close to rank column
16:30 jimallman agreed. so we’d trade the Show toggle for some kind of sort-by UI (sort by rank, sort by status)? i’d like to mock up a version that assumes displayed order is rank, and moves Excluded trees to the bottom of the list (no rank shown, no move widgets).
16:30 kcranstn is the plan that curators will be able to edit the descriptions and ranks from this window (i.e. the grey text)
16:30 jimallman yes, if they’re the owner or a collaborator, these fields would be editable. like the existing tree popup.
16:30 kcranstn ok
16:31 kcranstn agree about excluded trees not having a rank
16:31 kcranstn and like the sort-by UI
16:31 jimallman i’m thinking i need to add explicit Save/Cancel buttons here.. and make this a truly modal popup.
16:32 jimallman otherwise a casual/welcoming approach to collaboration might mean people drive by, play with some edits, and accidentally mess up someone else’s collection.
16:34 kcranstn or some sort of warning message about “edits here will be reflected for all collaborators”
16:34 kcranstn what about history?
16:35 jimallman hm, good question. i suppose i need to link to the github history, maybe even a diff of pending (conflicting) changes and a warning if they need to manually merge…
16:36 jimallman but yes, we at least need a friendly history as we show for studies. and a prompt for commit messages? that starts to feel kind of heavy for the task
16:36 kcranstn I wonder if a reasonable v 1.0 option is to not allow simultaneious editing? (user X is currently editing this collection)
16:36 jimallman i feel like commits to a tree collection are self-explanatory. or at least can be explained in the description field for each tree.. but maybe that’s a dangerous assumption.
16:36 kcranstn I think it is a reasonable assumption
16:38 jimallman hm, i’ll ponder what it would take to block simultaneous editing… we’ve already paid the price for phylesytem-style branches, etc. to handle conflicting edits, but that might trigger lots of manual cleanup. hard to say at this point..
16:38 kcranstn that might not be the best option, but worth considering
16:40 kcranstn I suspect that we are going to need to revise the collaboration aspect (owner gets to approve / reject? gets to invite collaborators vs public?)
16:40 jimallman so far we’ve talked about a pretty open model, where any logged-in user can “hop aboard” and make themselves a collaborator — or make a copy for themselves.
16:41 kcranstn but want to get something out for review before making those decisions
16:41 kcranstn yes, agree about that model for v 1.0
16:41 jimallman ok, will proceed on that assumption. if people want to be more protective of their collections, we’ll find out soon enough.
16:41 kcranstn indeed
16:42 kcranstn so need to think about what happens with merge conflicts, though
16:42 jimallman right. i hate to push people into git command-line, but it’s hard to see what we can offer short of that.
16:43 jimallman i think that’s our expectation for study conflicts, but these might be more frequent, and might happen to more casual users.
16:43 kcranstn yeah, getting worried about that
16:43 jimallman i suppose one option would be to make a copy in this case (in effect, their WIP branch is an owned copy)
16:44 jimallman and offer manual merge as an option for the truly motivated. though they might find it easier to simply repeat the same decisions in a chosen collection.
16:44 kcranstn “oops! someone else edited and saved this collection while you were editing! We’ve made a copy for you called “Collection title (Copy)” ‘
16:45 jimallman :)
16:45 kcranstn with link to Collection title history
16:45 * jimallman is thinking this trough...
16:45 kcranstn cool
16:45 jimallman s/trough/through
16:46 jimallman just like the over-stuffed icon bar, i’m watching to see if tree collections become too much to really handle in a popup…
16:46 jimallman with history, collaboration, “editorial” decision tools…
16:47 kcranstn I admit that I am not a fan of the popups (for trees, or for collections)
16:48 jimallman it’s an attempt to stay “in context”, rather than jumping away to a new page. at least for collections, this makes more sense, since they don’t “live” in any particular tree or study.
16:51 jimallman sorry, that was unclear. i meant that opening a separate page for collections makes more sense (to me) than for trees.
16:52 kcranstn makes sense
16:53 jimallman i’ll see if i can wrangle all this in a single-collection popup (w/ tabs?), and explore a full-page option if it starts getting too heavy.
16:54 jimallman other suggestions or concerns? if not, i have plenty to work with above. :)
16:55 kcranstn I think that’s it for now - other meeting coming up :)
16:55 jimallman cool, thanks again
17:02 jar286 joined #opentreeoflife
17:05 jimallman jar286: hi! fyi, kcranstn and i were just discussing changes and simplifications to the proposed collections UI, see the logs at http://irclog.perlgeek.de/opentreeoflife/2015-05-27#i_10664272
17:06 jimallman i’ll work on implementing some of these, and exploring others. thoughts welcome.
17:06 jar286 in meeting, let’s talk at 2 or so
17:08 jimallman no need, just FYI
19:12 jar286 joined #opentreeoflife
19:24 kcranstn joined #opentreeoflife
19:31 kcranstn joined #opentreeoflife
19:39 kcranstn joined #opentreeoflife
21:06 jar286 joined #opentreeoflife
23:05 jimallman_ joined #opentreeoflife

| Channels | #opentreeoflife index | Today | | Search | Google Search | Plain-Text | summary