Starbuck's dashboard

Bump - Still in the queue? Just noticed same issue today. Uh, no rush... :) – Starbuck, on November 12, 2009 01:46

Thanks Peter - I'll add a couple related notes here:

• I have a drop page up in blog view. I emailed a note to the drop and as expected, it "magically" appeared attached to another note. The two notes were flagged as "Today". I emailed another note a few minutes later, but the drop marked that as tomorrow's "Thursday" date. I switched to chronological view, then back, and all notes were grouped as "Wednesday", rather than "Today". The drop is set for timezone PST but I since it's now after midnight GMT I'm guessing email timestamps aren't accounting for the drop's timezone when they're first received.


• Again on the drop page, I have the chat section up, where I entered my name to replace GuestNNNN. When I change views, I re-join the page as a unique guest. It would be nice if the session state remembered the chat name.


• Related: When changing from blog view to chronological, the chat re-activates all recent notices, showing "The ___ 'text' was added" for all recent asset additions, so I see each notice twice.


• About the last two issues: Going from chronological to blog view, the chat messages are "undoubled" and the user name does not revert to GuestNNNN.

HTH, regards as always,
-S-

I dunno if this fits with the topic request, but I noticed that the RSS, when viewed in the browser, presents a redundant link: The subject of a given asset is a link, and at the bottom of the text is a link directly to the asset. I'm not sure if most RSS readers get that double link but it might be worthwhile to consider making that bottom link optional.

I just wrote a utility to copy or move assets from drops to an FTP site, and similarly to move files from a directory in an FTP site to a drop. I'm not quite sure what to do with it yet so suggestions for specific functionality are welcome.

An application I can see for this is a background process that watches drops for new files and then pushes them off to a desired FTP site. If the site contains files for a website, this means you can essentially update images and other content for a website simply by posting files to your drop. Of course that also means you can email files to your drop, which then get pulled out to the website, or you can have voicemail and other assets displayed on some website other than drop.io, you can even get inbound faxes posted to some other site.

Conversely, let's say someone posts a file to an FTP site. This will pull it out to drop.io, and from there rules can be applied for emailing, tweeting, or performing other standard Drop.io operations on the data.

This is one step away from a similar utility where you just put files into a local directory to have them pushed out to a drop - or similarly anything that gets put into a drop can get extracted back to your PC.

This utility is currently a Windows Forms (thick client) app written in C#. It can and will easily be re-worked to a tray app, a windows service, or incorporated into any other application. As a Windows app, this needs to be running somewhere in order to work. (Um, duh?) It can get run from the Task Scheduler, run in the background, or it can be triggered by some website operation. If you want the functionality to work without your PC running, the utility will need to run on a hosted Windows box somewhere. I can host this if there is demand, but as always (with me anyway), such a service can't be provided for free - maybe a buck a month or something close. I don't see a need to port it to Mono for Mac or *nix any time soon but if anyone really prefers to have this running outside of Windows I don't anticipate any issues.

Comments and suggestions are welcome.

I appreciate the thought process but there's a usability issue when someone looks at a UI and can't figure out what to do without being told.

I'm not looking for a new feature. I want to point people to a web page and get them interacting with it without me having to point them to a separate page of documentation.

Try this usability experiment. Create a drop with a couple assets. Add a comment to one asset but not to another. Now ask some people who have never seen drop.io before to comment on the assets in the UI. See how long it takes for people to figure things out without you telling them where everything is. If you need to tell the user anything, the test fails. If it takes them more than (arguably) 5 seconds of mousing around, it fails. If they can't find help to tell them what to do, it fails. If you consistently point people to a page and they "get it" without any wasted time or extra instructions, then I'll have to re-think about how much faith I put in the average website user.

The bottom line is that no UI should have too many hidden ways to access primary features. In some ways I think drop.io was designed to be a little too clever - designed for "us", not for "them".

I'd love to see people post results of the above test here, and please include some indication about the type of person doing the testing: younger/older? developer? internet savvy? complete noob? occasional surfer? ...

Ben - I had lengthy exchanges with Drop.io management about this topic as well. After a while I came to the conclusion that it may be better to give full credit to Drop.io and acknowledge them as providing key components of a value-add infrastructure. These days is sort of in-vogue to associate one brand with another, which is why so many sites have API links to Twitter or Facebook, or Google widgets and other such plugins. That's the nature of mashups.

I have no idea who you are, what your main product or model is, or what your goals are, so I hope none of this offends you.

If you intend to provide the exact same services as Drop.io, with no value add, then your clients don't have a lot to gain from coming to you versus going direct. If that's the case you're going to have a hard time with the whole venture. If you want to sell voice mail services (without the branded "welcome to Drop.io" message") or provide a way for people to share files that they email to a host (not @drop.io), then you should probably look elsewhere.

However, if you do provide unique value-add and the Drop.io services are merely a part of what your clients/visitors get when they come to your site, then I don't think you have to worry about people wanting to cut out the middleman or go directly to the source.

So my advice is to consider embracing the brand, and figure out a way to make people feel good about coming to your site to make use of Drop.io services. Provide some way to make use of the files and transport tools that compel people to come to your site - and if they go direct to Drop.io, many/most will realize that they simply can't get there what you offer. If people do go direct and find what they want, let them go. Competing with free and low-cost services is tough enough as it is. It's even more difficult to try to make a profit in the process.

Was just looking for the same thing... And I'll add here as well that it would be nice to have a Sticky option. I just created a drop by posting content first, then I decided it would be good to have a Welcome message to let people know what the drop was about. Oops. So I deleted the assets and added a Welcome note. But in blog view it falls to the bottom anyway. Double oops. What I want is a sticky welcome, then an admin default and guest option to toggle LIFO or FIFO under the stickies. Sticky items should remain at the top even if changing view from media to blog to chronological.

Thanks as always!

I'm seeing the same thing in v1.6.13 over Windows.
- Copy from an external source.
- Open a label, ctrl-v or use the context menu and Paste.
You don't see the text in the box but if you press Enter (or click away from the control, you'll see the text actually did get pasted in there ... multiple times if you kept hitting ctrl-v because it didn't seem to work. ;)

This seems to be a good place to make a note of ctrl-z / undo issues.

Ctrl-z as Undo, and ctrl-y to Redo, only work at the control level, not with text.

Enter text into any control, then ctrl-z. There doesn't seem to be a way to undo text being entered "right now". I know the next question is "undo to where?" What do you think about undoing text back to the last punctuation, or since the last carriage return? This sort of thing even works in browser text areas. Right now I'm tempted to create text for listboxes and accordions in a separate editor and then paste the final version into Balsamiq. That's just silly.

Same goes for ctrl-y, which of course if you can't ctrl-z you can't ctrl-y either...

As another example, I prefer to accentuate data rather than the labels which precede them. The default 'Label / String of Text' BMML is bolded which means I need to use the default once, duplicate it, modify it, then use my version rather than the Balsamiq control selector from that point forward. That's sort of a waste... I also add Search buttons next to many fields. Balsamiq has Help buttons but no search buttons. The list can go on, and of course every designer/developer has different preferences.

I think it's very important that the software allows users to add and remove controls from the control selector. The only way I can think of to easily use custom controls now is to create a BMML file of frequently used controls (oh dear, what an awful acronym) and then copy/paste them as required.

An alternative to modifying the top selector bar would be to add another selector at the top where you allow a shortcut filter for Common, Containers, Layout, etc. Add "Custom". Then when that item is clicked, import the BMML from all files in a defined directory-1. When we click on that item in the control selector, paste in the BMML from a file with the same name in directory-2. The reason for this is that there is only so much space in the control selector, so we can use dir-1 for "thumbnail" BMML and have the real BMML, maybe downloaded from Mockupstogo.net, in dir-2.