scirate priorities

1 view
Skip to first unread message

Ben Toner

unread,
May 5, 2014, 7:15:02 AM5/5/14
to scirate-dev, Steve Flammia
Hi Aram, Steve,

Now that the new scirate deployment is stable, can you please give rough priorities on how important the following things are? Feel free to add/change things too.

Also can you attempt to draw a line when you'll be comfortable enough with the site that we can all start promoting it?

* (In progress) Login from google
* Fix date selector widget, e.g., make it smaller, fix or remove "Since last opened"
* Better mobile experience
* Improve/redo design of site
* User profiles
* Include #spnetwork comments
* Email alerts
* Copy bibtex data to clipboard

Cheers

ben


Ben Toner
Founder & CEO, Draftable Pty Ltd

Ben Toner

unread,
May 5, 2014, 7:43:15 AM5/5/14
to Steven Flammia, scirate-dev
Thanks Steve,

I think the most important thing before we really promote SciRate to a broader audience is the inclusion of a recommendation engine. Jonathan Oppenheim and I were supposed to work on that… but you can imagine that without any pressure on us we didn’t do anything yet. 

I think it's the other way round: we need more users before a recommendation engine is useful.

I admit though that I am still confused about what problem the recommendation engine is solving. Are there some notes somewhere? I looked at this old planning document https://docs.google.com/document/d/1UtqoOKpl28lW8hdEklV4OzM9LX0cA13UV1fTxf3BCHE/edit?usp=sharing but it doesn't really explain.

ben

Ben Toner
Founder & CEO, Draftable Pty Ltd


On Mon, May 5, 2014 at 9:20 PM, Steven Flammia <sfla...@gmail.com> wrote:
Hi Ben,

My rough ordering for these would be:

1 (In progress) Login from google
2 Better mobile experience

3 Copy bibtex data to clipboard
4 Fix date selector widget, e.g., make it smaller, fix or remove "Since last opened"
5 Improve/redo design of site
6 User profiles
7 Include #spnetwork comments
8 Email alerts

Regarding #2, the chief problem is just that the left-most column doesn’t shrink into a tab on a mobile device. I really use bibtex copy a lot, so 3 is important to me. 

I think the most important thing before we really promote SciRate to a broader audience is the inclusion of a recommendation engine. Jonathan Oppenheim and I were supposed to work on that… but you can imagine that without any pressure on us we didn’t do anything yet. 

-- Steve

aram harrow

unread,
May 6, 2014, 10:10:26 PM5/6/14
to Ben Toner, Steven Flammia, scirate-dev
Steve, did you reply just to Ben on this thread?

Ben, I think the main reason for the recommender is the sense that
this site might not work as well for multiple communities, including
those outside of q. info, if (a) raw scite counts include lots from
people with different interests, and (b) similarly recent comments are
mostly irrelevant. But I could be wrong, and it's true that more data
would also be helpful in designing recommenders.

aram

Noon Silk

unread,
May 6, 2014, 10:38:14 PM5/6/14
to aram harrow, Ben Toner, Steven Flammia, scirate-dev
Regarding not worknig well for multiple communities; I'm not sure what you mean.

I note that if you are not logged in, scirate displays the most scited papers, regardless of their category. But if you are signed in, you only see the top scited papers in your chosen categories; so then you arguably want to see the most scited papers in those categories, right?

I agree that the comments might get hard to follow. Maybe that little bar for the "recent comments" could also be filtered to only the comments on papers in categories that you follow (but then there'd need to be another way to see 'all comments'.)
Reply all
Reply to author
Forward
0 new messages