Category Archives: technology

How to change the future

Business Insider ran this article today that caught my eye:

5 time travel questions to ask Google 

The idea is suppose you end up 10 years into the future, what would you ask Google before you came back?

This got me intrigued, beyond the obvious avarice inducing questions which presume biological and physical determinism that I reject because I believe in free will, what would you ask that could change the future?

And what I realized is that if you look at science, there is about a twenty year time horizon between an idea being valuable and an idea turning into a commercialized success. And so what I could do to change the future would be to find out what was novel in the future 10 year future, and bring it back thus moving humanity 10 years in some disciplines.

The question is which disciplines and where to get that information?

If I could bring backward some idea, in what discipline would it be?

The first practical challenge would be that I know a tiny sliver of human thought… and I am aware of a tiny amount of real science… So if I was going to time travel, before I got into my device, I would do some research about where to go to ask the questions. Then armed with those disciplines and and places to ask I would go ask.

Now suppose I found myself thrust forward in time due to a rift in the space-time continuum and I had only a few moments, I suppose I would ask

  1. Top ten breakthroughs in physics 2023
  2. Top ten breakthroughs in medicine 2023
  3. Top papers in computer science
  4. Key note speech siggraph
  5. top y-combinator 2023

quickly copy the results and head back to the past, and hopefully, somewhere somehow someone could use that data to move the human race 10 years faster than anticipated…


zomg – ux fail at smugmug

I have been a loyal smugmug user since 2004. I am such an old user of smugmug that when they did their big reveal they reached out to me twice to see if I would come to their HQ for the big reveal.

I was touched. And very excited. For years they have been the best place to store and view pictures. And I have happily spent 50$ a year for the privilege of having them store my pictures…

The process to get to the new UX … which looks a lot like the windows 8 UX… was

  1. Preview the new look and feel
  2. Update your page to the new look and feel.

So here’s the dialog that asks if you want to upgrade


Looking at this UI, there appears to be a progress bar, and two buttons.

Because I am an engineer who doesn’t RTFM, I click on the button ‘UNVEIL IT’ expecting the progress bar to start advancing…


And I get this red box. Which is surprising. I think… huh, something must be wrong at smugmug. Or this is going to take a while. So I let my computer sit overnight waiting for this progress bar to advance.

Fortunately my brilliant wife does RTFM, something about doing research I think, and points out that I must type the word unveil  in the text box.


I do that and get the new UX.

So yes, I am a fool for not reading the instructions. But a great UI has to not rely on the user to read the instructions. Or when the user fails to read them provide the right kind of information so they can take corrective action.

I wonder how many people are spamming customer support.



Time to invent a file system

In 2006, while at NetApp, I remember with horror the launch of Isilon.

Isilon’s product was everything Clustered ONTAP – aka GX – wanted to be.

One of the most intriguing aspects of the product was the use of Reed-Solomon codes to cut the amount of storage required. The downside, of course, was that rebuild was a bitch. The rebuild was so painful, that although the tech was interesting, our senior most architects were dismissive of the value.

They believed that the clustered storage solution and a clustered file system would deliver superior availability with better cost and faster rebuilds. Or something like that, I must admit that I have forgotten the details of the debates and don’t feel like pulling remembering everything.

The market failure of Reed Solomon codes, more or less convinced me that the right answer for the foreseeable future was 2x the storage costs.

And then I read this:

That is a nice summary of this paper:

This is a huge result. What it suggests is that storage availability is no longer tied  to 2x the storage infrastructure without taking an unacceptable hit on recovery.

A new file system that embraces this kind of encoding could be a good solution for a large class of applications that don’t need the RTO of 2x the storage. Making storage cheaper has always been a winning strategy for growing market share.

A new clustered file system built around this kind of erasure code or even a variety of erasure codes could be a significant new addition to the tech eco-system.

I wonder if something built ground up would look very different from adapting an existing system.


Winning the long game, Microsoft, Nokia, and Windows 8.1

2008_nov_windows_1_0The release of Windows 8.0 was a bold statement about the future that I agree with. The future of computing is touch screen devices with optional keyboards. And that an operating system that can make both work will win.

At some level, there is a large group of smart folks who disagree with the idea that the square can be circled, that the future is discrete distinct devices with keyboards dying a slow miserable death.

The challenge is that the majority of work is data entry. A keyboard is used for most data entry. And the most efficient typing device is a mechanical keyboard.

So the keyboard will continue to have a place in the market.

In this future, an operating system that allows both touch and keyboard data entry allows application developers to decrease their R&D. Instead of trying to build two distinct applications one for touch and for keyboard, they can think of touch and keyboard as two distinct views into their same underlying application.

And it is that reduction in R&D that will make keyboard + touch screen devices win out. If you make it more efficient to build  solutions, then the cheapest solutions to build tend to win out over the long term. And if you are Microsoft you can burn through cash to win in the long haul (BING!)

And that brings me to Windows 8.0. Windows 8.0 sucked and was awesome at the same time. Windows 8.0 was awesome because it absolutely nailed some of the frustrations around windows and app discovery and it definitely got me wishing for a touch screen on my laptop. Windows 8.0 was horrible because there were so many distinct usability flaws. For example, the fact you had to use the keyboard and the mouse to find an app, the annoyingly difficult ability to get the search icon, and I could go on.

Windows 8.1 is an incremental improvement.

And that got me thinking about Windows 1.0. I am certain when Steve Jobs saw Windows 1.0 he thought: nothing to fear here. And I am certain the UNIX guys saw Windows 1.0 and said: Nothing to see here. And then Windows 2.0 shipped, and still nothing changed. And then Windows 3.0 and it almost got usable. And then Windows 3.1 and the world finally tilted in Microsoft’s favor.

I have a strong belief in the value of incremental improvement winning out over magical product discovery. And Microsoft has always nailed incremental product improvement when they are moving in the right general direction.

The improvements in Windows 8.1 are noticeable. Is it a great product? No. But it took Microsoft 7 years to build Windows 3.1 and it took them 15 years to get to Windows XP – and that was the first version of the OS that actually worked.

So what can get in the way?

The real challenge for Microsoft is not that the path they are on is wrong. The real challenge is that from 1985 to 2000 Microsoft was the destination for the best and the brightest in the tech industry. The question is whether they can continue to attract the best and the brightest who can build that transformation…

Not dead yet.

In 2007 I bet that Nokia could figure out this iOS thing. And I was wrong. Nokia spectacularly failed to recognize the disruptive nature of iOS, sat on their lead and is now trying to tell us that they are not dead yet. I figured that with all of those resources, a competent CEO, a competent CTO and a strong technical team would seize the moment and realize like the British did with the Dreadnought that everything had suddenly changed and their lead had evaporated. And, without a shadow of a doubt, their CEO was incompetent and their technical team for all of Nokia’s incredible technical talent was unable to react to the iPhone.

Success or failure, ultimately is a function of being able to attract talent, point them in the right direction and have the ability to course correct over time. For Microsoft, the direction is right, the ability to course correct was demonstrated, now all that remains is whether they can attract and retain the talent to win.


bing! I want – bing! to – bing! – tell you – bing! something

One of the most irritating aspects of the Android user experience is the bing-bing-bing of notifications.

When you buy a new phone, it downloads all the apps you had in the past. Those apps have default configurations that cause notifications to be put in your notification inbox and to make noises.

And so you play this game of whack-a-mole killing the sound notifications only to have it go bing! at the most inopportune time because some app you forgot about decided that it was the time to go bing! as part of some vain re-activation play.

I wish, wish there was a global way to control these frigging notifications globally…

How ATT and Verizon turned mobile software into the new growth industry

Yesterday, I had the misery of replacing of my phone.

When I originally bought my Nokia cell phone, I bought it with the expectation that I would get an early upgrade in about a year. Given that the phone was free – thanks to a bug in the OS – the theory was that in one year I would buy a better phone at a reasonably discounted price to replace my aging and dying Nokia E71

Except ATT, and now Verizon, changed the rules of the game and the early discount no longer applied.

Which sucked for me… but it really sucks if you are hardware manufacturer…

In the hardware business, you really need to get people to upgrade to the next model to make money. If they don’t upgrade, then you make less money. It’s really that simple.

Storage as a hardware business is awesome because storage is like a consumable… as long as you are creating data you are buying more storage. Once you use storage it’s no longer re-usable for another purpose without deleting data and data never gets deleted.

But to upgrade to a new cell phone you need a compelling value proposition at a reasonable price.

With the early discount model the carriers had turned what could have been a two-year upgrade cycle into a one year upgrade cycle. And had allowed more share of wallet to go from software vendors to hardware vendors… Folks who could spend were spending more on hardware over two years than on software.

The carriers have now reversed that policy which means that the prices after one year have gone up.

This is really unfortunate for hardware manufacturers. To deal with the sudden increase in price, hardware vendors must either make the product more valuable through faster innovation, or figure out how to make the hardware cheaper or accept slower growth.

Changing the rate of innovation is hard. In fact, I would almost argue is impossible. Hardware rate of innovation is ultimately tied to Moore’s law. So you have to cut prices which is also sucks because it affects margins. Or you accept slower growth which isn’t so bad … other than the part where your shareholders ask you to do things like hand over your cash hoard and demand a dividend … and let’s not get into the employee retention thing.

All this means less profit for hardware vendors which means less innovation etc…

A slower hardware innovation adoption rate, however, is fantastic for software vendors. Unlike hardware, software vendors are able to continuously add value to devices without an upgrade cycle. In addition,for subscription services as long as the vendor adds incremental value the average sales price doesn’t have to drop …

So what does this mean?

If you consider the amount of money that the median first world person has as fixed or slightly declining over the next 10 years, then software vendors can capture a bigger share of the wallet.

Let’s be very specific:

Suppose a customer is will to spend 600$ very two years on a phone. In the old world, the customer could buy a 200$ one year, and a 400$ the next. With the new policies the customer spends 200$ and 600$ the next year because there is no discount. So the customer – unless he sees a compelling value proposition decides to not buy the 600$ phone which frees up 400$. That 400$ is available to spend on incremental software services on his phone. Although it’s certainly true that not all that money will go to software, but some of it will. And the really cool piece of news is that folks who were buying early upgrades have enough disposable income to actually want to buy more software services to extend the value of the device they already own.

If I was a software service vendor like Evernote or Google or Microsoft this is the best piece of news I have heard in a very long time… More money to spend on services.

If I was a hardware vendor this would suck. And the market agrees which explains the collapse of Apple shares.

And if I was Microsoft trying to grow my platform this would also suck because

  1. It means that growth of Windows Mobile will be slower as it takes longer for people to buy new phones.
  2. Given the investment consumers are making  in services, the stickiness of  incumbent platforms may increase over time.

My only hope if I am Microsoft is that they can somehow create faster software innovation that motivates people in the next upgrade cycle to switch … This is possible … in principle (bing is number #1 in search – right?) … 

And if I was blackberry, I would pray there were a lot of people who loved me…

And if I was anyone else trying to build a cell phone platform, I might be looking for a new strategy…

Nokia has chosen to die or The end of the Nokia affair.

Today is truly a sad day. After an almost 15+ year love affair with Nokia phones, I am moving on.

I feel lost and adrift, but it is time…

Nokia’s value prop was that the damned things never died and the audio quality is the best in the business. So in spite of their crappy UX, I loved the fact that my phone always worked, and provided perfect sound. In the early days of the iPhone (and to this day), I would get frustrated at the audio quality of their phones. Much like iPhone snobs hated my pictures, I hated hearing their muffled, distorted crappy voices.

But my Nokia 900 (edit: originally said 820 which is easy to replace) died a few weeks ago. And that is not okay. And it died because the internal and irreplaceable battery died. And I lost some pictures because I can’t get the pictures without tearing the damn thing apart … thankfully iFixit has ways of doing that.

I was thinking about buying another Nokia, but…

I found myself, before the phone broke, getting increasingly frustrated with the lack of apps, the poor quality of the photos, my inability to do things that my android device made easy.

And then I looked at the call log on my phone, and I realized that I barely used the phone as a phone. I have 14812 minutes on my cell phone. Seriously who need 14812 minutes on their phone (need to fix that bill at the next opportunity)…. That’s 10 days of continuous talking.

Heck I barely used it for SMS.

I mostly used it as a web-browsing device and an email device.

At work I have an Android device, and realized I was using this massive collection of apps that neatly integrated with each other (something the Apple iPhone experience works so HARD to make a frigging nightmare)… and found myself increasingly reaching for my work phone instead of my personal phone…

So good-bye dearest Nokia, Like many things in life, time has passed you by… Even this fan-boy bids you a not so fond farewell…

I never would have guessed that you would have pissed away such an amazing, and awesome strategic opportunity through colossal and spectacular business mismanagement.

As for you dearest Microsoft, love your desktop PC, love your mobile OS, won’t bet against you but you’re increasingly looking like OS/2 …



About f’ing time – Google admits GPA and Brain teasers aren’t useful

In life, it’s easy to say learn from mistakes, but it’s more powerful to learn from success. And successful people tend to learn from success because it’s so much harder to find.

Unfortunately for people like me, Google’s success has promulgated this mythology of the brain teaser question. Unfortunately for me, in spite of my professional success I find brain teasers difficult to answer under the pressure circumstances of an interview. I need time and space to think. And my approach to thinking is methodical and deliberate. This doesn’t mean I can’t handle  high pressure circumstances… Working at Zynga is one pressure packed cooker after another, it just means I don’t do brain teasers.

So it was delightful to read this…

What I really loved was this paragraph:

Forget brain-teasers. Focus on behavioral questions in interviews, rather than hypotheticals. Bock said it’s better to use questions like, “Give me an example of a time when you solved an analytically difficult problem.” He added: “The interesting thing about the behavioral interview is that when you ask somebody to speak to their own experience, and you drill into that, you get two kinds of information. One is you get to see how they actually interacted in a real-world situation, and the valuable ‘meta’ information you get about the candidate is a sense of what they consider to be difficult.”

What I learned in 1998 was that the best predictor of future success is past success. So if you want to understand whether someone is good at solving hard problems, ask them about hard problems they solved in the past and drill into that.

In a earlier blog post, and on quora I have given my dismissive opinion of the brain teaser.

I hope that the valley will now switch from asking questions that are brain teasers to questions that are meaningful and powerful.

It will take time for this approach to talent evaluation to change, but I know it will.


PRISM – and the dominoes fall

First journalists revealthat Mr. Snowden revealed nothing new. After all, we all knew that the NSA was monitoring everything.

Business Insider basically described Mr. Snowden as a dim-witted high-school dropout that caused himself a lot of pain over nada

And then we have this article that provides more insight into what PRISM does.

Essentially what PRISM does is give a narrow view into a small set of users that the NSA has legally obtained wiretaps that allows them to look at their online activity in real-time.

That is both easy and simple to do and does not need mythical amounts of infrastructure to do…

My original assessment still holds:the technological illiteracy of the average reporter, the rabid anti-government bias of too many folks in SillyValley allowed a low-level functionary to convince them that alien technology had suddenly materialized in the bowels of the US government.