The addition of Twitter and Facebook streams looks beautiful – I can browse (in a newspaper-style layout) the tweets, links, photos, and messages from both services quickly.

While I do love the layout (I’m a print guy, after all), I did find one problem with two areas of worry: The content from your Twitter and Facebook stream is presented in Flipboard’s own choice of layout. Meaning, some photos or links from your friends might show up as small boxes, or nearly fullscreen.

In the section view, this means any NSFW content uploaded by your friends (and we all have friends that post non-work safe images, or is that just me?) could be nearly fullscreen.
In the photo-slideshow view (which is displayed on-launch), that same content could appear fullscreen.

I use the slideshow view while my iPad is docked in my homemade cardboard dock at work, that way I can quickly glance at any given photo as it comes across and tap to read more. It’s a nice feature 99% of the time. That is, until a friend’s photo of a night out drinking or another’s link to a punk show (displaying an angry panda) might take some explaining to a hovering boss (as just happened to me).

Flipboard in the dock - showing the angry panda

Flipboard in my hand - showing a photographer-friend on a night out

Twitter stream - yes, that says 'warm weather humping' with a nude photo

As much as I’m enjoying Flipboard, I can’t recommend one using it as a work-safe slideshow or browsing app if anyone could be hovering in your work area.

That said, totally worth the wait for my custom access. I’m looking forward to the ability to add in news feeds in the future (hint at a request there, guys and gals), video and more.

I’ve written a plugin (for CodeIgniter) that handles the string as input, checks for empty lines (excessive \r\n’s), cleans e-mail addresses (pretty), and converts URL’s to their full ‘a href’ layout.

The problem I’d like to solve: many of the URL’s I’m finding are “long” – meaning, they are over 100 characters long. I’d like to shorten them to, say, the first 20-30 characters, followed by ellipsis (eg: “http://www.chadedge.com…”).

Using substr() inside a preg_replace() will not work (substr chokes or preg_replace chokes – neither like being wrapped up that way).

Any ideas?
–

Here’s a before:

its hard to find a good one and most of us are not to thrilled with some of the local repair shops. i have found a really great guy in north seattle. he does beautiful full restores, maintenance, rebuilds and great work all around. have been using him for a while (including correcting some of TL's work). He also sells NOS and repro parts for a better price then anyone local. the is the best i have found and really a great guy to work with. feel free to email me and i can tell you why i like NCVH better then TL. 

http://northcityvintagehonda.wordpress.com/

http://www.northcityvintagehonda.com/servlet/StoreFront

And the after:

its hard to find a good one and most of us are not to thrilled with some of the local repair shops. i have found a really great guy in north seattle. he does beautiful full restores, maintenance, rebuilds and great work all around. have been using him for a while (including correcting some of TL's work). He also sells NOS and repro parts for a better price then anyone local. the is the best i have found and really a great guy to work with. feel free to email me and i can tell you why i like NCVH better then TL. 

http://northcityvintage...
http://northcityvintage...

Update (2010-07-26): I have written a solution by editing (a bad idea, of course) the url helper (system/helpers/url_helper.php) in the function auto_link.
I highly recommend you do NOT repeat my process; however, as this is a one-time hack for a special circumstance, here’s what I did:

In url_helper.php – at the function_auto_link: There is a section of code that parses any found URL in to it’s “bits” (stripping http:// etc). I checked for the length of the url, and if it’s +20 characters, I substring it down to 20 and add an ellipsis to the string before returning the full linked-url chunk.

The code:

Again, I don’t recommend you follow my example – it’d be better to subclass the helper, but in my situation this works just jim-dandy.

For the most part, you’d be right. With one minor exception: corrupt fonts.

Mind you, this might be unique to me, but I offer this tip just in case you’re presented with the same issues I’ve faced.

First, the screenshot of iTunes shortly after my migration:

The solution I used to fix this font issue was to open Font Book (remember that app? Probably gets used as much as once a year I’d imagine). Launch Font Book and look for any fonts that have a yellow triangle (alert) next to them. CMD+click on the fonts and choose “Resolve Duplicates.” Or, you can just select-all and run the same command.

Quick as a whip, the problem was solved.

The only other issue I had with my migration was iTunes failed to see my music, movies, etc (and crashed quite a bit). A Software Update check fixed this (turns out, I was running 10.6.3 on my new machine, with an older version of iTunes, etc). Once I updated the OS and the available apps, all was running at top-speed.

Watson, I.B.M’s latest creation in the field of question-answering machines, is blowing my mind with every article, study, and video I can find. Here’s a promo video from I.B.M:

I’ve been researching a Bayesian matching system for comparing and binding news stories from different sources, with the goal of being able to match using only two sources (totally breaking away from the base logic of a Bayesian match algorithm). A question-answering machine may be much better and figuring out small matches in strings and titles where a pure keyword-based match would fail.

Let’s hope the Jeopardy! tests go well as I.B.M has suggested that Watson could emerge in business uses in the next couple of years.

PS: Here’s a longer article by the NY Times (which I read last night and tweeted about)

This is getting crazy.

What chaps my ass is when a designer/developer takes credit for a design that was not theirs – one that they have made adjustments to, while leaving the core design in-place. Further, when the core code lives on, with some adjustments, I also get a rash.

The case here is Muumuu Heaven, a Web site that a buddy of mine were hired to develop by an employer of ours. It was a rough time in our development lives – with difficulties in communication, slipping deadlines, design changes, underbidding, etc. A lot of the hardships were my fault, but in the end I was happy with the result. Not the best development in the world, but I applaud the designer I worked with for his color choices, images, etc.

It’s been a few years, the Web site has gone through some updates and changes – both in design elements and code, but much of my designer’s look/feel remain (meaning: the exact images he designed are still in-play; the scaffolding is the same); the code for navigating the site is the same.

On the footer of the Web site is a link for “Site Credits.” I had no supposition that Muumuu Heaven would credit the designer or myself, but I was surprised when the new designer and developer took credit for work that was (at best) half theirs.
This continued on the developer’s own Web site (which is where the following screenshot came from).

I’m not asking Muumuu Heaven to give us credit now, I’m just ranting about the ass-rash that things like this give me. I’ve contributed on many projects and do my best not to boast about being “the man.” If you’re going to claim ownership of a design or development, then make sure it really is YOUR design/development.

(end rant)

Of course, I had to ruin it by trying to cut a wake/sleep hole (dammit!), but beyond that: score!

Here’s some early photos:

And a brief video:

Details over the next month.

A 100% rewrite, tons faster, cleaner, and with some rich highlighting tweaks that I feel puts Searchy far ahead of other iPhone browsers.

But that’s all my opinion – what’s yours?

Apple Remote Desktop Issue from chadedge on Vimeo.

As you can see from the video above, when I connect to a certain Mac server using ARD, the screen “flashes” repeatedly – it appears that the ARD connection is dropped and resumed over and over again.

This issue only occurs with the specific server, has only started in the last week, and is repeatable on any desktop that uses the remote connection software.

I don’t believe any new software updates were pushed out to the server, and each desktop that connects is quite different, so I’m not sure where the problem lives.

using jQuery plugin “scrollTo” locks a UIWebview position from chadedge on Vimeo.

I’m working with the jQuery plugin “scrollTo” and I’ve come across a problem:

When I fire the $.scrollTo('.a_class_item',{duration:'800',axis:'y'}); the Web page I’m viewing on the iPhone does scroll correctly (very nice!).

However, once that first position has been scrolled to, control of the Web page is toast – meaning, if I try and touch-drag anywhere else, when I release my finger the Web page “snaps” back to the first scrolled postion.

I cannot duplicate this in any browser (in a desktop browser, the behavior is as expected, again, very nice interaction).

If I default back to a window.scrollTo(x,y);, then the Web page retains control and the UI responds correctly – it seems that the scrollTo plugin is “locking” the page position.

Any ideas on how to “release” that lock so I can scroll, resize, (even scroll to a different area on the page)?

I have a document that I’m traversing that has a “B” (bold) element, containing both text and an image “IMG.”

When I search the document, I’m not coming up with the text within the B as a match, and I’m wondering if it’s because the element.nodeType isn’t responding to type “3″ (which should be a text node) properly (as the text node also contains an image).

Any ideas on how I can traverse to find the text within that “B” as a matching string, excluding the img in my result?

(Note: I’m doing this as an update to searchy in order to make it more robust for both the iPhone and iPad).

While I don’t mind watching the film on DVD, I do like to watch movies while using my iPhone, and therefore the movies need to be seriously compressed.

I usually use VisualHub (now defunct, unfortunately), but sometimes dip into Compressor for the final encoding.

Today, I started to use Evom and I’m quite happy with the results.
While Evom did take much longer to re-compress my 1.6GB file (about 2 hours, compared to about 60-70 minutes with VisualHub), the resulting file size is a shocking 250MB less than what VisualHub gave me.

Here are the two compressed files, side-by-side. Can you tell which is VisualHub’s reduction and which is Evom’s?

Final note: if you CAN tell which is more compressed, do you really think it matters on the iPhone screen? I don’t.

Square has been on my radar for a few months – mostly for tinkering project ideas. Here’s some things I’ve been thinking:

1. Human ATM (swipe card, give cash) for places like Dick’s in Seattle
2. Buy tickets even easier from scalpers (ha!)
3. Sell your merchandise easier at live events (hello, OSNAP.net!)
4. Be the coolest lemonade stand on the block