My Twitter feature requests (for the Web interface):

• First-class support for tweet replies/comments
• Put the search box right on the page, upper right
• Let me add “searches to track” on my page, that scroll along in little widgets
• Provide a subtle visual hint (e.g. highlighting) where the tweet stack was the last time I visited

I’m sure these features exist in the various Twitter clients.  Twitter should be more aggressive about pulling in the best third-party innovations, as Google, Facebook, Amazon, and other platforms do.

(My friends know I can’t help myself from “product managing” others’ products).

Driving home today, the FastLane toll (E-ZPass for the rest of you) flashes “Call FastLane”.

So I call and am prompted for entering my account number “from my statement”.  There’s no option to bypass, but after three failed attempts, it gets through.   Then I’m put on hold for a while, and then eventually it says “we’re closed”.

Unbelievable.

Truly great designs are hard, but rookie design mistakes are usually easy to avoid.  In this case, most FastLane users will learn there’s an account problem when they get a “Call Us” flash at the toll lane.  As such, it’s reasonable to assume that many (if not most) users will be calling from the car, without statements and without account numbers.   Therefore,  an IVR menu that requires an account number is an incredibly poor design.

This stuff drives me nuts.  It’s so avoidable.

I thought Krugman’s Op-Ed about compensation in the financial sector was interesting.  The one-sentence summary:  “I question the value of these financial innovations, and why are they getting paid so much?”

Financial innovators have a huge advantage:  they build their compensation right into the innovation itself.  Find a place to take a percent or two (as risk-free as possible), and at scale you’ve got real money.

Compare this to (say) technology innovation:  getting paid is frequently as much work (if not more) than the original innovation.

Since it’s launch, I thought iPod Touch was underrated and under appreciated.  The iPhone’s got the glory, and it’s easy to dismiss the Touch as a “fancy iPod”.  But it’s a lot of computing in a small and cheap platform:  8-32GB of flash memory, 128Mb of RAM, a great display and a 522 Mhz processor.  It’s got more power than many PCs from 10-12 years ago!

It was interesting to read about the Army’s use of the Touch for solider applications. I think we’ll see the Touch in more cases where the economics of a general-purpose, mass-produced hardware platform enable an app that wouldn’t otherwise work.  Could you imagine the Army making a custom-hardware Arabic translation device?

Also, my prediction:  Apple’s “netbook” competitor will be a large-format Touch with Bluetooth support.  Add Apple’s little wireless keyboard, and you’ve got a fine laptop replacement for many users.

A classic “mobile app” company has a CEO or busdev person plugged into the carriers, a marketing person that knows about getting “on deck”, and a tech team facile in BREW and Symbian.

Now, with the iPhone, Android, and Palm Pre, those specialties are becoming less relevant.  These new platforms are more open and Web-based.  There are platform-specific elements, but existing HTML/CSS/JavaScript/Canvas development skills translate over just fine.

Also, many apps need access from both desktop and mobile screens.  Point an iPhone to any top Web site and you see functionality formatted and tuned for mobile users.  Even game developers (a key category on mobile) will want desktop Web access for player rankings, game info, new game content, gamer forums, etc.

The result of this convergence is that we’ll see fewer pure-play “mobile” companies, and more “just” app companies developing across desktop, mobile, and living room screens.  Old-line mobile companies will find their proprietary skills and relationships less relevant as things open up and become Web-based.

I’m wondering if in-flight Internet is going to lower demand for redeye flights, since you can get a “working day in the air”.

With the recession, inbound requests for “networking” have gone up quite a bit.  (If I’ve been hard to reach, I apologize:  networking time demands currently exceed supply).

I frequently get pings from senior folks, some hoping to find an opportunity at an early stage startup.  In many cases, the qualifying question is:  how long can you afford to work with no salary? Here’s why that’s key.

For many software technology projects, the initial costs are all salaries (especially these days).  More and more, entrepreneurs are scrapping it out and coming to the first investors with a prototype or beta, in some cases with initial market uptake.  By the time of first funding, the key roles are usually already covered by that founding team, especially for product or technology positions.  (Sales/marketing positions are one exception:  that leadership may come on board later, as the company grows).

In other words, senior folks looking for paid positions at early stage startups may find the train’s already left for key roles.

As we give up land-lines for cell phones, there’s still one useful purpose for the plain-old telephone:  providing a number for the entire “household”.

There are a bunch of cases where a household number is needed:  giving an emergency contact number for school/camp/etc., having a contact to give the power company/cable company/bank, etc.  And even mundane cases as having a number your kid’s friends can call, before they have cell phones of their own.

Google Voice is close to providing this, but seems more focused on the individual than the household.  I want a simple menu tree (“Press 1 for Andy, 2 for Kellie, etc.” that routes incoming calls to individual cell phones, inboxes, etc.  I want to transfer calls between cell phones, just like I transfer a call at home by having my daughter pick up the phone.

AND, this should be part of my cell phone provider’s “family plan” (I’d pay extra for it).  They could make it stupid-easy by porting existing analog lines when people sign up.

Why not?

Recent discussions about bailouts and regulations have got me thinking.

Corporations enable capitalism.  The main feature is limited liability, which limits owner’s losses to their original investment and employee’s to losing their jobs.  In other words, if a corporation has obligations it can’t fulfill, the owners (stockholders) and employees aren’t on the hook.  This feature lets investors risk capital without worrying about losing their homes.

Moreover, this liability limitation is unbounded — if a corporation racks up $1 or $100 trillion of obligations, a bankruptcy cleans it up and everyone walks away.  Unfortunately, as we’re learning, US and world economy does not have an infinite capacity to absorb losses.  When corporations amass obligations so large and far-reaching (e.g. AIG, LTCM, etc.), we can’t (collectively) afford to let them fail.

Now, we’re debating regulations to prevent this from happening again.  Regulation has a number of problems, including the ability of free markets finding a way to route around, inventing ever complex financial structures, and making money in unregulated ways.  Regulation may prevent this crisis from happening again, but won’t prevent the next one.

There is an entirely different way to address this, by targeting the root problem.  What if some personal liability phased in for cases of huge, “off the charts” liability?  In other words, if your corporation amasses obligations so large that bankruptcy would materially affect the US economy, you might not be able to keep your houses and bonuses.

I guarantee that a threat of the smallest personal liability will cause a whole lot of self-regulation.  The art would be to do it in a way that it only applies to the extreme 0.01% of cases, and untended consequences (e.g. the problems with Sarbox) are minimized.  It may be possible to make this work.

It’s a slightly crazy idea, but not entirely crazy.

With the proliferation of public computers (e.g. libraries, hotel business center, etc.), we need a browser with a lockable “public computer” mode, unlockable only by the computer administrator.

In this mode, no passwords are stored, and auto-complete for form fields is turned off (or quickly expired).  Any open windows would close automatically after some period of inactivity, and the browsing history is quickly expired.

Also, the browser would send this mode information in all Web requests (in an HTTP header), so apps could modify their behavior accordingly.  For example, a Web app might NOT offer the option to “stay logged in” when using a public computer, and/or might aggressively time out an inactive session.

I’m sure someone’s thought about this — is this a new idea?