Friday, September 18, 2015

The End of Profit

Finally I get to write my own "The End of..." piece, albeit just in my own blog. To not bore anyone with the requisite amount of text appropriate for the title, here's just the executive summary.

The never-ending quest of the business world to increase efficiency means increased commoditization of every product, with which margins become really thin, making it very difficult to turn any profit at smaller scale, which in turn would disincentivise new business formatiuon, which will depresses the economy and wages. Where did I make an error here?

Thursday, September 17, 2015

Amazon Fire TV 4K is Alright, Not Mind-Blowing

Amazon has pulled ahead from Google by releasing the 4K-capable version of the Fire TV mini-STB. Google's Chromecast does not yet have a 4K game. Until recently it was only $200 Nvidia Shield console ($225 with the remote you want) that supported 4K.

So the question for 4K folks is: Amazon Fire TV or Nvidia Shield? Or, even, why bother with dongles and set-top boxes instead of just using Smart TV capabilities of your 4K TV set?

(Again, I am not a gamer, I am interested in high-quality media streaming, so my take on this reflects only media part of the equation.)

Here's the things: YouTube, as of now, holds advantage of the 4K content available. In order to play YouTube 4K content, your system needs to have hardware support of VP9 codec - 4K format used by YouTube. No all 4K TV sets support it. For example, Vizio, which combines excellent features with affordability, can't play YouTube at 4K - a major drawback, IMO. However, a TV set that does run YouTube at 4K with comparable feature set, will cost you about $800 more. Now, to me this $800 gap for just a couple of features I miss, main of which is 4K YouTube, is the space where 4K STBs come in.

Until today, just about the only 4K STB game in town was Nvidia Shield console. But with today's announcement of Amazon Fire TV, we've got competition. And to me decision for which one to get comes down to this:
FeaturesNvidia ShieldAmazon New Fire Tv
4K Frame Rate60fps30fps
YouTube 4K SupportYes?? No

Yep, from its specs, it's unclear whether new Fire TV supports VP9 YouTube codec. Until I can confirm it's a "yes", for me Fire TV is not a contender. If the answer is yes, then I can imagine trading extra 30fps of Shield for a $100 discount.
According to, YouTube 4K content cannot be watched due to lack of VP9 codec support by the new Fire TV. So I $100 price tag seems too high for this device.

Wednesday, September 9, 2015

Entering 4K Video World: Hardware Upgrades and Process Changes

4K video resolution is arriving - not really slowly, but in a somewhat weird way: 4K content is sparse, while consumer-grade hardware to produce and consume 4K content is getting affordable, from GoPro 4 Black and Sony FDR-AX33 camcorders costing under a $1000, to plenty of sub-$1,000 4K TV sets. This price point is a sweet spot where a video geek can get the hardware and not be overrun with guilt that must ruin the psyche of a real early adopter.

Watching raw 4K footage on a 4K TV is not something I can deal with - video has to be edited first, and in the 4K video case, it means your computer will have to move and process a whole lot more data. Which in turn means upgrades, and no matter how exciting it is for you and me, it means keeping costs down to make this all bearable for our dear sane family members.

Start with a monitor: your HD computer monitor will need to be upgraded to see the glory of 4K. I got Acer S277HK. It's not perfect: backlight is bleeding over at the corners pretty badly, but it's 4K, it's IPS, it has HDMI 2.0 in addition to the DisplayPort so it can be connected not only to a computer in a pinch, it has built-in speakers so I don' have to waste desk space for speakers, and it's not ridiculously expensive. I am very happy with it, running on Windows 10. Everything look massively more gorgeous on it than on a regular HD monitor.

To drive the 4K monitor, I got the least expensive video card I could find for the purpose, which is just about any GeForce 750Ti - it won't handle 4K gaming, but should be plenty for 4K video editing. I got the EVGA, which worked out quite nicely. It also has a fairy quite fan, and I really like when my system is quiet. GeForce 750 (non-Ti) might have worked, but based on reviews it looked iffy, so I went with the Ti, which seems to be very much up to the task.

Now, here's something I didn't experience in a while: after copying 100 Mbit/s 4K footage shot on Sony FDR-AX33 from an SD card on to the 2TB spindle SATA drive, I found that playing back 4K 100Mbit video from the spindle is all jittery as heck. I felt it is 90's all over again. Now, how do you fight this? The largest 4K clip I have is about 46 Gig and I definitely will have lots of these in the future. A reasonably priced large-capacity SSD that would provide adequate performance still costs about $250 on sale. At only 0.9 TB, it won't take very long to fill it up with 4K clips. So I got an idea: replace the spindle drive with the SSD, and to save the space, get Amazon Cloud Drive Unlimited. At $60 a year, and assuming the price will go down in the future, unlimited cloud storage looks very attractive for anyone planning to edit 4K videos. As an Amazon Prime user, I have been relying on complimentary Amazon Cloud to backup my pictures for a while. Using Amazon Cloud Drive by itself is not a particularly polished experience: one has to use awkward Amazon Cloud Client to upload files, which is a far cry from, say, DropBox quiet syncing. So although in theory Amazon Cloud Drive Unlimited seemed like a good option, I still wasn't sure how the whole workflow would look like, until I discovered the Odrive - the syncing client Amazon Cloud Drive sorely misses.

ODrive is the secret sauce that made this whole thing come together. Now all pieces were falling into places: upload all your raw footage to the Amazon Cloud Drive Unlimited and using Odrive, unsync folders with massive raw 4K footage to save the space on the SSD. When editing video, sync only folder(s) that hold 4K footage you are working with. Once done, unsync to free space. Rinse and repeat. This way your large-capacity SSD drive will hold only your pictures and the minimum number of big 4K video clips. That can easily be managed with probably even a 512GB SSD.

Now if initial syncing of the 4k raw footage up to the Amazon Cloud Drive won't run into the ridiculous and dreaded 300GB data cap of the blood-sucking Comcast (Update: oh yeah, I did - see update below), I should be all set for editing my 4K videos on a budget, using VideoStudio X8 Pro, and not having nervous breakdowns due to fears of running out of disk space or slow hard drive performance.

Finally, once I finish editing a few 4K videos, I will stop memorizing the content of the awesome and get an actual 4K TV to enjoy the fruits of my labors.

Update: ODrive Windows client crashes fairly regularly. If it crashes in the middle of uploading a large file, it starts over. It crashed several times trying (and never succeeding) to upload my 46GB 1hr+/100Mbit/s 4K clip, which took my Comcast data usage to 670GB, while I never reached the 300GB cap before. Comcast forgives first three months of overages under the "courtesy months" policy. I'll is that's the case when the bill shows up.

Wednesday, May 27, 2015

Sunday, May 3, 2015

The Search Is Over

Google found out that "When you look at people who don’t go to school and make their way in the world, those are exceptional human beings. And we should do everything we can to find those people."

Dear Google, don't kill yourself: the search is over, I am right here. You are welcome.

On Being Rational

I got no well-structured thoughts on the subject. For now it's just a list of points for future elaboration.

  1. Being rational (having common sense) means being persuadable by evidence. Everyone will agree with with, but most people will firmly hold on to their deeply-held irrational convictions.

  2. Majority of opinions we hold are prejudices, biases, and instilled dogmas, rather than tested assumptions that were proven to be correct. Our identities are largely based on these untested opinions. Parting ways with one's identity is an extremely traumatic experience, hence people often would rather die than admit they were spectacularly wrong, even in the face of incontrovertible evidence. Early-formed, quickly-formed and strongly-embraced identities are the surest way to becoming deeply irrational.

  3. Rationality can be made more wide-spread if it's treated as a form of intellectual hygiene: compensating for own biases and stamping out fallacies in ones thinking is not natural, but also not terribly difficult: it can be learned, and the habit of questioning own assumptions can be maintained through practice, like washing hands and brushing teeth. To that end, rhetoric and logic should be taught, with spotting fallacies being the most important skill practiced in lab exercises. Ability to detect untrustworthy sources of information is not a natural proclivity either because it requires ability to adjust for own biases. But that too is a reasonably easy skill to acquire.

  4. Historically, being rational was not the most advantageous behavior: proving that Earth is not at the center of the Universe could get you killed. Allegiance to the tribe was valued more than being right about facts. Repeating disproved theories as an article of faith is just a statement of allegiance to the tribe, to reassure identity, and to get protection and the status among peers. As tribes become less essential in providing security for individuals, mutants who practice rationality are free to use feedback loops for quickly iterating through trial and error in chasing larger status goals: having much larger scale of impact and acquiring more wealth.

  5. Most opinions stated on TVs are not backed up by anything, and usually are pile of fallacies.

  6. The most exciting word for a rational person is "because". Claim without proof means nothing.

  7. In the long run, incentives inevitably trump culture. Aspirations and convictions melt away as compliance with incentives get justified even if incentives contradict views dictated by the culture. Structuring incentives correctly is more important than efficient day-to-day management of performance problems.

  8. Being right in business was always more profitable, but few businesses until now were structured to separate clear performance data from the noise of subjective opinions about the contributor. Also, even if a business was successful vs the competition due to acting in a rational manner, until recently businesses didn't have the global reach to compel the competition to also adopt measuring performance. Current situation where self-measuring businesses get very decisive competitive advantage is the single largest contributing factor for increased rationality in the society in general.

  9. Big Data dramatically simplifies and reduces costs of searching for evidence.

  10. Problem: when measured criteria are too few and simplistic, incentives get narrowed down to focusing on passing the test at any cost, including taking on insane workloads and cheating. Another problem is negative motivational power of measuring one's performance: people get turned off if they fear they may score poorly, leading to the vicious cycle of "low performance => poor score => even lower performance".

  11. Cooperation is more profitable than confrontation. It's obvious on the surface, but in evolutionary terms, humans had such a small amount of resources throughout the history, that someone's benefit almost certainly was somebody's loss, hence zero-sum mentality took hold and still dominates (sports, wars and border conflicts lasting well into the era of potential abundance). That's why it looks like humans programmed not to truly enjoy their success unless competitor loses. Most successful societies are those where trust covers widest possible number of people. That's not really natural, as evolution taught us to trust members only inside small social network: a family, a tribe, ethnic group, etc. That's why corrupt and brutal but familiar local leaders are routinely elected and stay in power as long as they can maintain the fear of the larger groups of outsiders even if the leader produces dismal outcomes for the group.

  12. Adding value (working) has become more profitable than taking property of others (war) very recently in historical terms. That's another facet of the reasoning why humans are ultimately held down by using zero-sum thinking as a default route.

  13. From evolutionary perspective, status among peers is the only important measure of success because it lead to reproductive advantage. Hence physiologically status is more important than wealth because wealth is only a proxy for status. Since status competition is zero-sum by definition, the question is "how much money does one really need" is moot: winner will be satisfied with one penny as long as everyone else got nothing.
If you agree with what I wrote above, you are in danger of being irrational as I made lots of claims without providing any proof or references. :-)

A Horrible, Terrible, No Good, Just World Is Coming. Thanks, Big Data.

The arc of history certaily bends towards rationality. Big Data revolution is making sure that few decisions will have to be made based on pure opinions and hunches, a.k.a. prejudices, biases and dogmas, which are also known as bad assumptions.

Isn't that exciting that in the fast-approaching evidence-based future incompetent bosses will be quickly found out despite their self-confidence and charisma, full-of-it pundits will be exposed, hot-air-filled politicos will be voted out, and humble, talented and under-appreciated working bees will be promoted and rewarded accordingly?

The answer is easy, it's yes, which is why "Moneyball" is such a bestseller: a wiz-kid mined data to find a factor in player stats that has strongest yet non-obvious correlation with winning. The justice is served to multiple parties: lone wiz kid with proven track record has become an existential threat to the army of highly-paid scouts producing dubious output, the open-minded GM saves his career in the big way, and talented underdog players took their righteous place under the spotlight. Just world, here it comes, and although the road to a giant digital BS filter is going to be bumpy and winding, to me it's pretty clear we are heading in that general direction, and going there pretty fast.

I am also sure* that this just, rational world will turn out to be extremely unpleasant and disappointing for maybe the majority of people. See, right now, while working as a mid-level office drone or a retail sales clerk, or a barista, we all enjoy plausible deniability of being simply under-appreciated and thus we can maintain our self-esteem knowing than some stuck-up/suck-up got ahead simply because life is not fare and bad things happen to good people - us. Now, what if Big Data revolution shows that not only our mediocre bosses do not deserve telling us what to do, but we too turn out to be just as unremarkable as them? What if it turns out that there are people, who are so much more talented that they can add much more value than just about any one of us? What if majority of population is getting paid right now to simply slow down these remarkable individuals and drown their weak, nerdy but brilliant voices in the noise of millions of opinions? Well, even I turn out to be that brilliant individual, who is going to be shown objectively better at doing my job, and many others will have to rush implementing my genius ideas, I fear most people will be so permanently and irreparably depressed that I won't be able to enjoy my new-found appreciation :-). Worse is, of course, if my track record, studied under the microscope of Big Data, turns out to be just "meh" and I will have to leave behind ranks of creators and idea generator and take the spot in the swelling crowd of implementors and assistants, or worse, to become a stipend receiver paid to stay home and not break anything.

As they say, the factory of the future will have only two employees, a man and a dog: the man will be there to feed the dog, and the dog will be there to keep the man from touching the equipment. The only question is are there enough dogs?

*This post will surely one day be mined and ranked by our robot overlords for how well my prediction worked. If I turn out to be right, I hope they will consider me for the dog feeder position :-).

Tuesday, April 28, 2015

IaaS is Not a Real Cloud

In my not-so-humble opinion, putting your client-server app on an AWS or Azure VM is no more "Cloud" solution than running DynamoDB and telling people you've got "Big Data".

To me, real Cloud and real Big Data start with using scalable PaaS OS, like Azure or AWS. If you got no PaaS, don't talk cloud and big data.

Learning Azure on Your Own Dime - Extremely Expensive

OK, I am completely sold on The Cloud, and specifically on PaaS. I am ready to go use it, especially Azure since I like Microsoft stack and tools, despite AWS being more popular and having more features (we, artisans ;-) develop stronger connection with our tools and techniques, than craftsmen or business people.) I started learning Azure, got the account, a VM, Azure SDK, developed an interesting testing-the-waters project - the Azure Storage Queue multicaster, similar to ESB topics.

My first Azure bill, despite MSDN subscription was above $150. I run a 24/7 application server on a slow VM there, where majority of the cost comes from, but small experiments, like with the VPN for example, added $25 before I noticed that it's really expensive. That was unpleasant.

Now, I want to experiment with something other than what I can run on local Azure simulator (Tables, Blobs and Queues). I would like to play with ESB, API Management, DocumentDB and just about everything else, but I am really concerned about how much cost I am going to incur by using each part of the stack, especially if I forget to power everything down on Azure for a night. The point is, my impression is that simple poking around and learning Azure stuff is rife with unpleasant and costly surprises. All these risks and impediments combine into a single big turn off for a potential Azure developer.

With Azure lagging in adoption behind AWS, slapping down even dedicated fans like myself, means that Microsoft is very unlikely to ever catch up with AWS. "Developers! Developers! Developers!" indeed. I know it is possible to plan the cost ahead using complex Excel spreadsheet, and there is an almost-meaningless 30-day free trial (learn new enterprise OS in 30 days, really??), but all these impediments add up to one huge fogget-bout-it!

Here's what Microsoft could do to make it easy for me to learn Azure: create closed-off sand-box on Azure infrastructure, inaccessible from the outside, but make all Azure stack technologies and components available there for free to all MSDN subscribers. Let us play with and learn Azure for free, and charge us only when our wares is made available public. That would be a huge differentiator for Azure.

If you agree with the statement above, talk at Microsoft about this.

But while Microsoft is ignoring us, I decided to do my part, in the upcoming weeks and months will do public service so to speak, and will pay out of my pocket for learning Azure, and will blog about my experience, including costs. I will tag those posts with "azure" and "paas". Follow this blog to take pleasure in my future misfortunes on the path of conquering Azure.

Monday, April 27, 2015

Photography is Easy

The truth is, good photography, although not exactly trivial, still requires a lot less skills than, say, painting, or playing a musical instrument. Anyone with a modest doze of persistence can figure out basic relationships of ISO, aperture and shutter speed, effects of different focal length, white balance and tone curve. Then all it takes to make a few good pictures is to take thousands of whatever pictures. Here is the result of my cherry-picking a few shots I mined from my digital collection. Most of them shot on the modest Canon PowerShot S95 and lightly edited in Picasa.

On a PC, hit F11 to put your browser to full-screen and then click on the picture to see it in the album.

Hakone, Japan

Rain in Wyoming

Paris, of course

Seattle Skyline


Caminito, Buenos Aires


Key Largo, FL


Kauai, HI

Napali Coast