Bryan finally posted something about the Journal idea that he and Seth had been worked on way back when. We’re using a lot of these ideas for some of the basic interactions in OLPC. His mockup was rudimentary, as most initial mockups should be, but I thought it would be worth it to post one of Eben + Pentagram’s most recent mockups to give people a better sense of the direction we’re going.
Note the critical pieces that are included: the document itself, who you worked on it with, name and metadata in the form of tags and the ability to easily share it with other people. Also note that we’re using “Resume” instead of open. The idea here being that you can actually restart an activity in the context of the people that you were originally working with on the document. So it’s not just about about the “what” of the data but also the “who” and the “when” each of which is missing from our current desktop metaphors.
This is the one part of OLPC and Sugar that needs the most amount of work and is the highest risk. But it’s probably the most useful part of the entire user interface. Ever notice how any reasonably complex cell phone or embedded device includes a file manager? There’s one on the N800, there was one on my old cell phone and they were the most confusing things. What do you do with data when you have it? Where did it come from? How do you give it to someone else?
The Journal also solves a few other interesting problems for us. The first is backups. The laptops don’t have a lot of storage, so it’s important that backups are done on a regular basis. So how do you go find old stuff when you need to? In our case, because the journal is time based if you’re close to the server your old stuff can just “show up” in the journal. So backups become offline storage instead of being a huge copy of everything you have. (Of course, you can backup what’s on your machine too, but it’s just how you view your data that matters.)
The second problem is “what do you throw away?” When you’re running out of space on your machine you often get a warning that says “hey, you should delete something.” But what do you have that you can delete? The Journal makes it easy. You just delete old stuff that you haven’t starred as important. (The question of whether or not that’s done automatically is another question, but we’re flexible enough to go either way.) It also makes it possible to make trade-offs. That is, if you want to download this thing that’s 50MB in size you can say “OK, these are the 5 things that you will need to remove to make space – is that OK?” Simply adding time and a simple value (i.e. if it’s important) means you can make all kinds of value decisions on the things you carry around with you. That’s very difficult to do with a standard desktop model.

It is good to try something different than open but the word choices “Resume” and “Erase” do not feel quite right to me. Cannot quite put my finger on it but perhaps this cosmetic detail will change in later iterations.
The journal idea sounds very intuitive and if the sense of importance is balanced right, and there is an easy way to judge against the largest files when it comes to clearing space.
I am not sure how well the ‘Web 2.0′ technologies such as tag will work well with young kids. Personally, I don’t think they have such many documents to warrant the need of ‘organizing’ documents. Adding them only adds more confusion.
For me, I think the best representation would be visual. A simple screenshot and description, and an easy way to browse through a gallery of documents should be sufficient.
> Simply adding time and a simple value (i.e. if it’s important) means you
> can make all kinds of value decisions on the things you carry around with
> you. That’s very difficult to do with a standard desktop model.
For time on the desktop, we already have mtime/ctime/atime.
Hey, I think this is a pretty cool way of doing ‘work flow’ ignore the negative stuff people post. They are part of the ‘change is bad’ monkeys.
Have you let any kids play with these machines for an extended period of time? My friends 6yr old spends hours on messenger with her friends playing flash games on ‘tinernet. I think chat + simple games is a winner.
monk.e.boy
I agree with a lot of the thought behind the journal but the mock up interface seems insanely complex and confusing, even for me. There are a lot of buttons with unclear meaning
Will kids care about tags at all? Many adult users don’t care for them and I don’t think kids will understand or appreciate the concept of using as few unique tags as possible for the best organization. What tags will kids use? Every drawing will have new tags if they have tags like “ocean”, “fish”, and “bubbles”.
Isn’t the picture going to be like 1 or 2 inches tall if it’s using that little space on the small olpc screen?
Even storing multiple versions of a file is going to confuse the heck out of kids… it’s a very abstract thing to say that there’s this document and this document… and both of them are sort of the same document but they’re also sorta different.
Like I said the philosophy behind journals is pretty sound from a technical perspective but it’s going to have to be a lot better implemented to be of any practical value on the kid’s desk.
Pingback: One Laptop Per Child News
One question about backups:
-If I deleted some old stuff that I marked as not important, but my friend did not delete it, will I see it on my journal when I get near him? In other words, are you also implementing some kind of distributed, decentralized backup?
one question about versions:
-How do you keep track of all versions? Do you save it all? If in the next drawing session “Rika” is not available, when she comes back and resume her work will she see what her friends draw without her? Do you keep ALL versions?
and one comment:
-Love it all. But don’t you think you are going too far with rounded corners? Gimme some 90º!!
Tim Reilly : Which buttons exactly could you not figure out? they are all very clear to me. Other than the “share with [CD/circle/group/world]” they all are very clear
Frank Ch. Eigler: Having the modification/creation date available is very different than have it easily acessible for use. The fact that you CAN write a smart folder that “selects all files and folder older than six months” is not the same thing as basing the directory structure around it.
Cheng Khoon: Most kids I know have 4-5 classes everyday not including homework or games they do on their out of school time. That would beover a hundred new “activities” per month, considering that each tasks has only one “file”. Tags, in that sense are a great way to categorize them. Maybe changing the name to “categories” or “this work relates to those ideas” and putting some automatic tags might be better. But tags are better than, for example, renaming directories.
Alan: Erase and Resume are better than delete and open because that’s what you do in the real world. You don’t delete or open an unfinished drawing. There might be better choices than those, I agree, but remember that no kid will be using the sugar UI in english anyway, those are just a base for the international translators. The real OLPC will be in portuguese, nepali, thai and spanish.
I agree with some of the other comments that tags are going to be a fairly difficult concept to convey, and a hassle to create. By the time folks realize they should have created them they already have a big mess of uncategorized history and data that they won’t want to touch but aren’t sure they can delete (like my inbox would get before I read GTD).
I like the idea of automatically assigning some metadata. Obviously, you will get the file creation, last modified time, activity, who you were collaborating with, etc. for free. So what else could you use? Maybe there are other cues that could trigger memory. Weather… hot, raining, smoggy. If the child ever indicates mood in some way, perhaps that. Any kind of holidays might be relevant search criteria. Certainly if there are date searches, you need to make it easy to pick a specific month, and relative terms like “last week” or “last monday” or “after the first exam.”
The idea of a “class schedule” and syllabus leaps to mind. If the day is organized in such a structured way, you could infer by the time of day a reasonable tag to assign. If the year is divided meaningfully, before first mid term, after, etc. that might help.
Likely what is needed is some first-class organizing priciple above the level of “activity.” Maybe Problem Domain, or Task. Say you are writing a report on Volcanoes. You will probably start with a web search. Then you might send an IM to your friend to see what she has discovered. You could send a question to your teacher about when the report is due. All of the communication and research (texbook markup/ web bookmarks) probably belongs under the VolcanoReport tag.
So where does the tag come from? Perhaps the teacher assigns it at the beginning of the project, or you get a collection of these when you receive the electronic syllabus? Say we’re working on a science lab now. So all laptops get distributed a tag for the project, maybe it is text or maybe it is a picture. The child clicks on it, and it becomes a default tag for artifacts from ALL activities until that tag is closed. One could certainly imagine multiple tags being in force at once.
By the way, it has always struck me as strange that “Activity” seems to map directly to an Application. In human language terms, an activity is not an application. An activity at least spans applications. But if the ship has sailed on that term, fine. Call it Task or Effort or Investigation, or Exploration.
The bottom line seems to be, context either chosen by the child, given or shared by another student, or assigned by the teacher needs to be made easy to use and share. The tags need to become first class modes in the system. You are always in the Exploration tag, say, and you get a few default tags that you can pop over exploration (at home, at school, happy, sad, bored, …)
If tags are easily exchanged, granted/assigned, and created, kids will use them. If they feel like some geeky programmer oriented concept only the budding programmers will appreciate them.