Pushed 0.26 with markdown support. There are a few known bugs, like “code” mode mixes “pre” and “code” tags, interacting poorly with bootstrap. There’s no good syntax highlighting. But we’ll figure it out. 

Also fixed awful bugs, like every entry having a “test” tag by default. Began oauth integration, but have yet to push the oauth consumer keys. I really want to get mendeley integration and github integration up and running over thanksiving! 

Mendeley is the new hotness in citation management, and they have a great API. I’m really excited to use them, and to suck in my entire existing document collection. The downside, however, is that their “get metadata for this entry” service still needs a lot of work. Perhaps I’ll spend a little bit of time today writing a better one? It seems like a reasonable API-exercise. 

I had hoped this would be worthy of an 0.25 release title, but you know, weekends have a way of slipping away from you. So we’re calling this as 0.23. Major new features: 

• Support for API keys. Which means that now, you could potentially call directly into the mementa API. This would let you do things like, say, have a matlab command to insert figures as new entries. 
• Tags for entries: Now you can tag entries! Somewhat of a pain without entry-tag-autocomplete working, but “small steps” 
• This is a big one : We are now live at http://www.mementa.org !  

Minor bug fixes: 

• Editing an entry no longer takes you back to the top of the page
• autogrow for text entries works again

Some bug fixes and new features: 

• Pasting a url into the text entry auto-links it
• Fixed some horrible drag-and-drop, server-retry bugs
• Initial attempt to create entry permalinks! 

Released 0.20 tonight, with three new main features: 

• Drag-and-drop uploading of PDF, PNG, JPEG as figures works
• Reordering of entries works
• can now “archive” notebooks

as well as a host of other minor stuff including

• Fixed the footer-overlay bug
• should look better with narrow browsers (i hate you, CSS)
• Math/LaTeX in figure captions works
• Avatar tooltips
• Nasty tag indexing bug
• “pre” button for code editin

We’re going live Monday with :

• multi-user collaboration
• multiple notebooks per user
• live wysiwyg text editing with integrated LaTeX support
• tagging and organizational tools

and maybe a surprise or two :) As always, a lot of issues left to tackle. Quite a fun weekend ahead.

Here are the current issues:

https://github.com/ericmjonas/mementa/issues

Follow @MementaNotebook on twitter if you want an invite code!

I’ve spent the past two weeks of off time rewriting the page rendering logic to use a state machine for updates, edits, and transitions. At long last, this should enable the following: 

• multiuser editing of entries
• multiuser editing of pages
• notifications of conflicting edits
• unification of a lot of the code

Look for it soon! 

I really need to remember that the MVP is the minimum viable product : you have to be able to play with it. So I’m pushing my MVP until Tuesday. Yay, release slippage. Things I need to have working, though: 

• Page metadata : For each page, polish up the view of “most recent edit”, etc. 
• Finding stuff: If you can’t easily find the pages, this isn’t going to be of much use, is it? So I need to get at least rudimentary searching working
• List all entries / list all pages, and sort them

I’m keeping track of these bugs via Github Issues for MVP 1 . Let me know what you think? 

Say you want to WYSIWYG edit text on the internets. There are the stalwarts, TinyMCE and CKEditor, which work with a hidden iframe and basically seek to emulate as much of MS Word as possible. There’s also YUI2’s Rich Text editor, but I never quite can bring myself to suck in the entire YUI2 widget framework. 

Then there are the next-generation editors, which potentially use contentEditable. My favorites thus far are; 

• Aloha : Great, super-active project. 
• Proper: “semantic” HTML rich text editor

I’ve been playing with TinyMCE, and it’s really well documented, but still feels like a very pre-web2.0 project. It’s bloated, and there’s a bug in chrome where sometimes it creates a FREAKING HUGE edit window. 

What do I want out of my text editor? 

1. The ability to do reasonable, WYSIWYG rich-text editing, with support for bold, italic, underline, strikethrough, lists, and (important) headings and preformatted text blocks
2. The ability to have multiple editors active simultaneously
3. “feel” fast. Real quantitative, I know
4. Not make me want to hate life
5. Be supported in FF/Chrome

That’s not too much to ask, right? 

The goals for Mementa MVP v1.0 (with the remainder of my vacation) 

• Multiple users
• Deploy via dotcloud
• WYSIWYG editing of text entries (no latex yet, no imaging yet. Link to other pages)
• Simple always-entry-order viewing
• Page data: edit title
• List all pages with title
• mostly focus on single-user behavior

I’d also really like to have it be not ass-ugly. Why, CSS, why are you so hard?