Starbuck's dashboard

The more of these inquiries I see for support for various "otherware" the more I think it would be better for Balsamiq to have a published API, or simply good documentation for integration - maybe just a couple more options in the command-line. The FOSS market is pretty good about doing its own integration when there is something to plug into.

As I understand it, the basic functions require Balsamiq to do the following:
- Accept request for New mockup
- Accept existing filename
- Save BMML/PNG to specific paths and perhaps filenames provided at launch

If the response to that is "there are more things that need to be done", then just document it - that becomes the beginning of an API.

From there it would be up to product-specific developers to add their own mechanisms to supply Balsamiq with what it needs during launch and to embed results in their UI when Balsamiq closes.

Personally I'd prefer to see more development on the core product than on extensions into other products which are subject to break as soon as some of those other products have a new release or change their API, etc. Let each team work on their own specialty.

But if you do continue down this path, plesase add to the list : WordPress, Simple Machines Forum, and Mantis bug tracker. LOL

I've read recently in web pages and books on good design that Lorem Ipsum can be a bad idea. This short blog entry has an excellent alternative idea, and it links to another very popular page at 37signals.com, talking about why Lorem Ipsum is bad. I'm no purist on the topic but I think the advice should always at least be considered.

I dunno if Peldi would be interested in adding that into the product but as a developer I can easily see an external extension where you select a BMML file, tell it to change specific controls like that. If the file is open you'd get a notice that the file was changed and you can simply OK to reload and see the new controls. I don't think Balsamiq is "extensible" so I don't think guys like us on the outside can do this as an addon to the product. But if you don't mind a separate executable then I don't think this would be too tough.

Any thoughts from anyone about similar extensions or other "gee I wish I could zap my mockup" ideas?

Peldi, I have ideas of how to allow Balsamiq to be extensible without many changes on your side. Interested?

Peldi, is your vision of this to have a licensed version which people can host on their own sites?

I was just doing a presentation yesterday for a bunch of people at a company that work in different offices. I suggested that they could use Balsamiq and send me their final mockups for coding. I immediately thought they would need an extranet/hosted version so that they could collaborate. As it is now with the AIR version they would need to have the software installed on one or more workstations and use GoToMeeting or similar to collaborate.

I'd love to have a version on my website where I can get clients to login via my own interface (we're basically a .NET house here) and then allow them to modify/save mockups. Ideally I would attempt to secure mockups via the login ID to prevent one client from accessing someone else's mockups.

You wouldn't want the software to run away with someone publishing open access (like the limited demo published on the product site), so I'd expect/understand some limitations, perhaps mandatory branding inserted into the bottom of every BMML identifying the licensee. That's a sensitive issue and I don't have many ideas at the moment.

Ciao

Peldi wrote: "P.S.No, I do not want to offer users a way to choose their font in the UI, it's not needed and most fonts look like crap. You can choose a font face through the config file."

I think I was in agreement with you on all points until I saw this. On one hand I personally think Comic Sans is a bit over the top. On the other hand, I wrote about this in my blog and got an emphatic response that a mockup needs to make it apparent that it is not the finished product - and the font really helps with that.

In a presentation situation where we don't know the viewer(s) sense of aesthetics, I think it would be valuable, if not prudent, to have an option to quickly get out of Comic Sans and into another font. This should not be the focus of a presentation and we shouldn't need to say "wait a minute", go change the config, then re-open the mockup.

For these reasons I think Comic Sans can remain the default, but I actually think a font changer in the UI would be a good enhancement. Let the viewers decide how crappy different fonts are, and allow us to find something they like quickly. Our task is to get beyond this issue as quickly as possible and get them to buy into what we're really trying to demonstrate.

Sorry. :)

The FUSE interface would facilitate that. http://drop.io/clorinda This was (is?) being created by Diego Cena, one of the other guys out here in the world of people who like to plugin to Drop.io via the API. Unfortunately that project is for unix-like systems and won't help Windows users. Though it is possible to do a FUSE interface to Drop.io for Windows users too.

As I see it, with FUSE you can open a file locally and it would set a comment in the drop that acts as a lock. If someone else tries to open the asset the same way they won't be able to. Write the "file" and the asset is deleted (archived for versioning) and re-created.

The interface I'm writing now allows two people to maintain a file simultaneously and asynchronously - it's much less work than a file system interface. For most environments that's OK because one person would be working on a doc in response to someone else. There's rarely a good reason for two people to be modifying a doc at the same time. If two people do update the file at the same time, the last one wins - but a backup will be kept of all updates.

Another approach to this, perhaps more elegant, would be to use cvs/subversion to modify your docs and then have the respository sync'd with Drop.io. Those tools are already made to do what you want, and Drop.io would function, as designed, as a database of sorts rather than a workflow manager.

Again, all of this is possible. Is there any company out there who can fund the development? The way this and all other markets work is simple:
- If you want a feature you can ask the provider (Drop.io).
- Wait for their decision, and possibly for them to implement the feature. They may do it or they may not. You may get what you want or you may not. They may charge for new functionality or they may not.
- You can also wait for someone in the open market to do this for free in their free time.
- You can write it yourself.
- You can commission someone else to do it. A company can commission a developer to write the code (find people in the API forum). Or a group of people can offer something (cash, pizza, beer, dog food) as incentive for someone to take the time to create a custom solution. Hmm, that's something that could be managed with a drop like the survey: http://drop.io/starbucksurvey001

I'm not leading in any particular direction here, just making sure everyone is aware of the various options that are available to make use of the fine service provided by Drop.io. – Starbuck, on January 22, 2009 21:00

Zoho is pretty complete on its own. With Zoho Creator I can see people creating custom apps to store and retrieve data with a drop, dish out custom phone numbers for voice mail and conference, send faxes, maybe even playing back voice mail from within a Zoho app.

As always, these things are possible but someone needs to be motivated to write and support the code. – Starbuck, on January 21, 2009 21:00

I just had a look at the Diigo API, and yes, it's certainly possible to allow exchanges of bookmarks between Diigo and Drop.io. I dunno how much demand there would be for that though.

If anyone is interested, I just wrote a utility the other day that watches where you're surfing with Internet Explorer and it can upload your entire history or a filtered subset of URLs to a drop. For some people that sort of utility can be highly valuable - for others it could be a nightmare. The same utility could be made to allow you to select a block of text (sort of like Diigo) and send it off to a drop rather than just doing a copy/paste locally.

When it comes down to it, almost anything is possible with these tools, but frankly it's tough to justify spending any time on cool utilities without some sort of business model. Heck, I might post another survey on my drop called "Would you pay a buck for this...?"

This is an old topic but I was recently asked about security. Without looking too deep we came to the conclusion that companies should not be using drop.io as it is currently implemented if they need to be compliant with HIPAA, Sarbanes-Oxley, SEC, or similar regulations. We were also considering file exchanges among teachers for a large school board but I really don't want to deal with constant "what if" and "are you sure" questions. The bottom line is that drop.io is a great service but probably should not be used to store data that will change someone's life if it's lost or compromised. If you're concerned about security, encrypt your files before transmission, password-protect the files, password-protect your drop, and if you're really paranoid, use a private server and not a shared host. – Starbuck, on January 14, 2009 19:57

See this other note:
http://getsatisfaction.com/dropio/top...
In short, I'm syncing a directory on my system with a drop, though not using the dropFS code that someone else had written,

Because this is a background project I have to work on it only as time permits amongst others. So far it looks pretty good though and I'll probably post screenshots on my blog soon.

Why isn't it just a slam/dunk? Well, as Sam said, it's dangerous stuff, but that usually doesn't stop me. ;) Let's say I edit a file locally and it gets uploaded to the drop. Your system sees the update and downloads it, so you make changes and save it, thus causing it to get pushed back up to the drop and back down to my system. What if I'm still working on it? What if it's a large file, do we keep pushing it up and down from the drop? How often do we ping-pong files without abusing resources? Do we keep backups on the drop and if so, how many for any given file? How do we avoid blowing our MB limit?

As you can see there are some things that need to be thought through, but for a simple "all I want to do is save my stuff" solution, I should have something within the next couple weeks.

I agree that getting the feature from Drop.io would be good, but there's no telling when that might happen.

I'm almost done with my utility that creates a mirror between my hard drive and a drop. That means instant backups and instant document syncing with friends, doing nothing more than using files on your local system just like you always do. More info to come...

LOL - Sam, I've been talking with Jake for a long time about "branding" ... same concept.

This could be done with an external program that doesn't yet exist but it could be created pretty quickly: Drop a file to a folder on your PC and it will automatically replace an older version of another file by the same name in your drop.

I haven't looked at this but the only problem I would have with writing that would be that Drop.io renames files as they get converted, and I'm not sure we have a way to know what the original file name was so that we can do the swap. I could attach a comment to file/assets with the original file name and maybe some other "metadata" about updates.

Does anyone have a better idea?

I think a Note should be added to converted files which includes all available metadata, including the original filename, ID3 fields where available, and any other information which can be extracted from various data tagging formats.

The problem I see is that this puts a burden on Drop.io to understand and provide for whatever format people declare as being "essential".

A better to approach this might be to provide a mechanism that allows files to be processed by whatever tag extraction utility people find useful. I dunno if Drop.io internally prefers .NET or Java or some other tools, but consider this: An MP3 file is dropped and is just about to be converted. The drop Admin has specified that they want MP3 files to be pre-processed with the ABC data extraction utility. So the MP3 gets processed with that utility and is then converted as normal to the standard drop format, and all metadata is available in an attached Note as XML or maybe "Key:Value" fields.

Drop.io would need to solicit contributions, and certify submitted pre-processors - and this probably means getting the source, and recompiling themselves to ensure integrity. If you find a better way to extract ID3 or other data, try to hook up the author with the Drop.io guys to get the utility certified.

Another approach is for someone here to point to some utility that Drop.io users can use to extract metadata before they drop new assets. I can easily envision a program that monitors directories for new files and extracts data from the files into another directory. So all you need to do is drop the file in this special directory first, then drop it to Drop.io, and you've got all the ID3 info without Drop.io having to do anything. You should be able to add the extracted into to your own drop as a Note or Comment.

Hmmm, I think I could write that... I'll look into it.

What do the guys at Drop.io think?

I think this has to do with the right click context menu, which for some drops would be a good idea but most of the time I'm finding it gets in the way. You either need to provide more options to make it more versatile, or just add three options in the mean time:
1) Allow admins to turn it off from the drop admin panel.
2) Allow users to turn it off from the context menu itself.
3) Allow admins the ability to allow visitors to turn it off or not.

Have a look here:
http://getsatisfaction.com/dropio/top...
Note the link to drop.io/hacks