Actually, while writing the roadmap for 1.1 and daydreaming about XWatch's future, I realized, once again, that Watch is not a feed reader. Potentially an article (let's call it "item") could be created from any source: any webpage (see http://jira.xwiki.org/jira/browse/XWATCH-162 ), written by the user from it's own sources (by filling in title, content, description) and, ideally, created from almost any other sources (office documents, pdf documents, emails, various markup formats (newsML, docbooks, etc), feel-free-to-imagine-whatever) such that XWatch becomes a complete watching tool. Because of this, we could redirect the data sources interface towards a more "sources management" approach rather than "feed reading approach". WDYT? Particularly, marketers & client project managers: what do you think about this approach for Watch? (_complete_ collaborative watch tool vs _rss feeds_ based collaborative watch tool)

> Hi Cati and devs, > > sorry for the late response, here are some of my thoughts: > * First of all, I particularly like the solution you found with the > in-place forms (the add feed form, the tags, comments, etc). Keep > in mind > that, if needed, we can devise a way of disabling one or more of the > panels (feeds panel, articles or filters) to simulate a modal > dialog with > an in-place form, so feel free to dream on about this kind of > solutions, > even if you need modal behaviour. > * Second, I have some observations to make regarding the > scalability of > your proposal: > - keep in mind that it needs to be perfectly internationalizable: > all > the text should be displayable in different languages and should > fit (I > am thinking here at the tab names in the filters panel, for example, > about the small space you would have to fit the add-edit feed / add- > edit > group forms). By the way, are you thinking about keeping the current > design with fixed column sizes (200px side columns) and article panel > expandable or are you thinking about percent scaling, à la Google > reader? > - the tags and keywords in the text analysis can be any word and > potentially any size, so I seriously doubt they will fit properly > in 2 > columns as you printed them (via Marta) > * Third, you should study the possibility of keeping icons > consistency > with XWiki (e.g. edit, delete icons) (via Marta) > * Speaking of icons, I have a personal problem with the "go to > original > article" icon. It's really hard to tell what it represents > (although it is > suggestive enough, once you figure it out) and it's very hard to > describe > in a manual / guide / tutorial, otherwise then showing it. > * Potentially, at one point, there will be some actions that only an > administrator of XWatch could do: articles archiving setup, > parameters > setup (no. of articles in the list, no. of tags in the tag cloud, > etc), > and others. Besides this, client customizations would potentially > need > such admin actions too. I would like to see your opinion about an > administration panel/placeholder in your interface proposal (that > would > contain some activities and would easily admit other activities to be > added "transparently"). In particular, where do you see this kind of > "control panel" placed? In the reader interface or in a wiki page, > accessible from both the reader interface and the portal? > > * For article "deletion": we (me and Cati and Guillaume) had a > somewhat > long fight on the chat about naming this action "delete" or not. > Currently, the name of this action is "trash" and it is more of a > "vote-against" than a delete: it is the complementary action of > "flag" and > its result is that the article is no longer displayed in the list by > default, this must be explicitly requested by the user from the > filters. > But the article (wiki doc & object containing the data) is not at all > deleted from the wiki. In contrast, the feeds deletion, groups > deletion, > keywords deletion do delete the corresponding objects from the > wiki. I > would prefer a "vote-against" like name for the article action, > rather > than a name that would suggest deletion. WDYT? As a side note, we > could > think about integrating the xwiki trash bin in watch, to have > recoverable > data. > > Cati, thanks for your work > xwikiers, thanks for your feedback > > Happy coding, > Anca Luca > >> Hi, >> >> The new XWatch User Interface Proposal is available at: >> >> http://watch.xwiki.org/xwiki/bin/view/Design/NewUIProposal >> >> I'd be glad to get some feedback either on the list or in comments >> right >> on >> the page. >> I would also like to thank Anca, Guillaume and Eduard for their >> feedback >> and >> help given so far :) >> >> Thanks, >> Ecaterina Valica

You mean like this? http://n2.nabble.com/XWiki-Watch---Support-for-non-RSS--tp516351p516355.html

On Thu, Jul 17, 2008 at 9:42 AM, Vincent Massol <vincent(a)massol.net> wrote: _______________________________________________ devs mailing list devs(a)xwiki.org http://lists.xwiki.org/mailman/listinfo/devs

Hi, I've read the proposal and I like it in general. Nice work! <roadmap/general ideas> Stepping back a little and talking about roadmaps I think we should eat more our own dogfood, meaning we should use XWatch internally for our needs and find out what's useful for us first. This would be a good process to define what's useful for others too. Personally I'm not using any of the Watch UI right now. All I use is the resulting RSS feed (this one: http://watch.xwiki.com/xwiki/bin/view/WatchCode/PressReviewRss?space=XWikiS…) I don't know how others in XWiki are using XWatch but I have the feeling we could learn a lot from that.

I think there needs to be some incentive for users to use the tool (the filtering/categorizing part) they'll just use it as a basic RSS aggregator (as I do). I don't have the full answer to that but I can think of several ideas: 1) Its scope could be extended so that it supports different input sources (as discussed previously and also by Guillaume in this thread): a) Mails from mailing lists (this would require difference actions, like "closed discussion", "associate jira issue", etc. b) Documents in a WebDAV store c) XWiki Documents (pages) - This could be a nice way for people to classify/navigate in a wiki.

2) Integration with external tools. For example if I could flag articles directly from my favorite RSS feed readers without having to open XWiki Watch I'd definitely start tagging/filtering. I think this could be done easily by adding HTML at the beginning or end of feeds provided by XWiki Watch. We could add some Flag/Comment links there that would open inline (Javascript - I think most feed readers would support that).

</roadmap/general ideas> WDYT? Thanks -Vincent Note that solution 1) a), if done correctly could potentially solve our need for a forum. What I'd like right now from a forum is 2 things: the ability to extract some statistics (namely to list the top 10 contributors) and to tell when a thread is closed or not (it's not closed if the question asked has not been satisfactorily answered). I know we're getting a bit away from the initial Watch idea but I think it could broaden its usage and it would definitely fit in the "organize and categorize information" domain. On Jul 14, 2008, at 9:29 AM, ★Ecaterina Valica wrote: _______________________________________________ devs mailing list devs(a)xwiki.org http://lists.xwiki.org/mailman/listinfo/devs

On Jul 21, 2008, at 1:33 PM, ancapaula.luca(a)xwiki.com wrote: Definitely. We're talking about the same thing. I'm talking about it at the user level and you're talking about it at the implementation level :) I don't think this is the same thing. In that issue it says: "She goes to the XWatch reader and clicks on the "Add Web Bookmark button"". My point is to remove the need to go Watch and still be able to filter/ tag/comment.

Thanks -Vincent _______________________________________________ devs mailing list devs(a)xwiki.org http://lists.xwiki.org/mailman/listinfo/devs