Category Archives: News

Shared items and notes from my feeds and browsing. Subscribe as feed.

After years of insisting that DRM in HTML wouldn’t block open source implementations, Google says it won’t support open source implementations

Source: Boing Boing

Article note: Surprising no one, it turns out the DRM-in-HTML EME proponent's claim that it wouldn't block open/free browsers and web technologies was absolute bullshit, and the process was a sham enacted to secure power for entrenched interests.

The bitter, yearslong debate at the World Wide Web Consortium over a proposal to standardize DRM for web browsers included frequent assurances by the pro-DRM side (notably Google, whose Widevine DRM was in line to be the principal beneficiary) that this wouldn't affect the ability of free/open source authors to implement the standard.

The absurd figleaf used to justify this was a reference implementation of EME in open source that only worked on video that didn't have the DRM turned on. The only people this impressed were people who weren't paying attention or lacked the technical depth to understand that a tool that only works under conditions that are never seen in the real world was irrelevant to real-world conditions.

Now the real world has arrived, and it was just as predicted. Samuel Maddock is a free software developer who is creating a new browser called Metastream, derived from Chromium, the free/open version of Google's Chrome. Metastream is designed to allow users to "playback videos on the web, synchronized with other peers."

This is obviously not a copyright violation of any kind. Metastream allows users to stream copies of videos they are allowed to see, but synchronizes playback so they can watch them together. In an age of Twitch, this is obviously useful (also: it's something I personally ghost-wrote in to the BBC's 2006 Charter Renewal document as a favor to one of the people involved, so it's something that major rightsholder groups like the idea of, too).

Maddock wanted to allow his users to do this with the videos they pay to watch on Widevine-restricted services like Hulu and Netflix, so he applied to Google for a license to implement Widevine in his browser. Four months later, Google sent him a one-sentence reply: "I'm sorry but we're not supporting an open source solution like this" (apparently four months' delay wasn't enough time to hunt up a comma or a period).

The connection to the Article 13 debate should be obvious: for years, advocates for the Directive insisted that it could be implemented without filters, but of course it requires filters. Likewise, for year, EME's backers insisted that it wouldn't prevent us from having open, auditable, free-as-in-speech browsers that anyone could inspect, improve and distribute. But of course it does.

Of course it does.

I’m now only left with two options regarding the fate of Metastream: stop development of a desktop browser version, or pivot my project to a browser extension with reduced features. The latter requiring publishing to the Google Chrome Web Store which would further entrench the project into a Google walled garden.

I tried creating a web browser, and Google blocked me [Samuel Maddock]

Boy howdy, this is one subject where I loathe saying "I told you so", but... I sure told you so.[1] [Diaz/Hacker News]

(via Four Short Links)

Posted in News | Leave a comment

Google’s constant product shutdowns are damaging its brand

Source: Ars Technica

Article note: No shit.
An artist's rendering of Google's current reputation.

Enlarge / An artist's rendering of Google's current reputation. (credit: Aurich Lawson)

It's only April, and 2019 has already been an absolutely brutal year for Google's product portfolio. The Chromecast Audio was discontinued January 11. YouTube annotations were removed and deleted January 15. Google Fiber packed up and left a Fiber city on February 8. Android Things dropped IoT support on February 13. Google's laptop and tablet division was reportedly slashed on March 12. Google Allo shut down on March 13. The "Spotlight Stories" VR studio closed its doors on March 14. The goo.gl URL shortener was cut off from new users on March 30. Gmail's IFTTT support stopped working March 31.

And today, April 2, we're having a Google Funeral double-header: both Google+ (for consumers) and Google Inbox are being laid to rest. Later this year, Google Hangouts "Classic" will start to wind down, and somehow also scheduled for 2019 is Google Music's "migration" to YouTube Music, with the Google service being put on death row sometime afterward.

We are 91 days into the year, and so far, Google is racking up an unprecedented body count. If we just take the official shutdown dates that have already occurred in 2019, a Google-branded product, feature, or service has died, on average, about every nine days.

Read 33 remaining paragraphs | Comments

Posted in News | Leave a comment

Symbolics Lisp: Using the DEC Alpha as a Programmable Micro-Engine (1993) [pdf]

Source: Hacker News

Article note: Holy shit there is a lot of good stuff in here. I don't usually pay a lot of attention to the LISPers, but, from a 1993 paper: 1. Correctly predicting that microcoding was coming back to deal with the CPU-Memory speed difference. 2. Fuckin' cycle-counting performance instrumentation feedback IN THE TEXT EDITOR. I STILL WANT THAT SHIT.
Comments
Posted in News | Leave a comment

Against metrics: how measuring performance by numbers backfires

Source: Hacker News

Article note: Yup Yup Yup. I become more and more convinced that bureaucratization is the limiting factor of societies. Goodhart's law ("When a measure becomes a target, it ceases to be a good measure") guarantees that we can't effectively measure, and thus Pournelle's iron law of bureaucracy ("In any bureaucracy, the people devoted to the benefit of the bureaucracy itself always get in control and those dedicated to the goals the bureaucracy is supposed to accomplish have less and less influence, and sometimes are eliminated entirely.") takes over, and now you have a giant overhead machine that occupies all available resources. It's the "mythical man month" problem at maximum generality and scale.
Comments
Posted in News | Leave a comment

UK students on hunger strike want help for low-income peers

Source: Kentucky.com -- State

Article note: On one hand, doing _anything_ at UK is byzantine, so I absolutely believe the protesters claim that the services are hard to access, and I have no doubt that many of our students have basic needs not being met. On the other hand, there are already six full-time staff in 2 departments for this. The demonstrators are demanding another full-time position and another facility. I would be surprised if that didn't imply at least $500,000 a year in overhead on the task already, adding more people and more departments with overlapping duties but separate expenses will only increase overhead and obstruction. Make demands that actually help your goal.

A group of University of Kentucky students has started a hunger strike that they say won't end until the administration creates a "basic needs center" to help low-income students. The … Click to Continue »

Posted in News | Leave a comment

Ignore the noise about a scary hidden backdoor in Intel processors: It’s a fascinating debug port

Source: The Register

Article note: Everything is too complicated.

VISA: It's everywhere (on the system bus) you want to be

Researchers at the Black Hat Asia conference this week disclosed a previously unknown way to tap into the inner workings of Intel's chip hardware.…

Posted in News | Leave a comment

Biohacking Caffeine : Perfecting Dosage and Timing

Source: adafruit industries blog

Article note: Neat. A reasonably complete, deeply researched dosage scheduling tool for Caffeine.

Looking for ways to justify that fourth cup of coffee for the day? There are MANY studies on caffeine consumption including ones that look at fat burning enhancement, ideal dosage and timing. Let’s take a closer look at a 2018 study which we have pointed out before, but never took the time to explain how it works. The military department of defense and biotechnology have developed an algorithm that reduces our need for caffeine by 65% while increasing our alertness by 64%. How do we apply it to ourselves for optimal coding and focused hardware development?

The above screenshot was generated using the 2B-Alert Web page. This site was developed by the same military researcher who wrote up the study on “Caffeine dosing and strategies to optimize alertness during sleep loss“. Once you signup to create a free login it will ask you to set your unbelievable strong password. If you can get yourself through the password process it is on to typing in sleep data and caffeine dosages. The site provides automatic conversion from different caffeine sources so you don’t have to figure out how decipher milligrams on your own.

There are three graphs that can be generated once the sleep and caffeine data is entered. All of these are based on the Unified Model of Performance (UMP) which predicts effects of sleep loss and caffeine, as a function of time of day.

Mean Response Time – More alert means lower values. You can see above that each cup of coffee I consume creates a slight decrease in response time (faster, more alert). I found that less than four cups of strong 8oz aeropress coffee left gaps in my day where I got slower.

Mean Speed – This chart again shows clear peaks at coffee consumption times of 7am, 10am, 1pm and 3pm. My speed is projected to increase through the day. I tried generating graphs with only 1 cup and 2 cups of coffee a day, but it showed my performance was pretty much over by 4pm if I didn’t keep on chugging the black juice.

Lapses – The less lapses you have the better. Again the 4 cup model seemed to keep my lapses to a minimum for a 8 – 6pm work window.

While the 2B-Alert webpage is a bit clunky to use it does allow one to plugin exact sleep periods and specific caffeine doses for up to a week at a time. If you knew you were going to be in a sleep deprived situation and wanted to calculate the ideal caffeine consumption this would be the tool to use. My interest was more in daily performance to keep up a consistent work pace so my sleep schedule and coffee consumption look the same for each day.

Posted in News | Leave a comment

End-User Programming

Source: Hacker News

Article note: This is a topic I'm always conflicted and/or outside the major camps on. I do believe that systems should be designed for people to customize them to their needs, and develop both general and domain specific skills as they use them. The presence of the amazing things some people rig together with spreadsheets and other simple, familiar automation tools are plenty of evidence that this is a tenable system. On the other hand, I firmly believe the empirical evidence is that most people really don't have the developed reasoning skills to design and make things, and ultra-low-barrier-to-entry simplified systems only work to make dreck that they're designed to coerce. The process of working sophisticated relatively general tools is necessary to actually perform creative design. The rough conclusion is that we really need to design our social and educational structures to better support people onto sophisticated tools, rather than building coercive tools for idiots. That's a harder, longer, and less attractive to rent-seekers process. It's also vastly more empowering, and necessary for computer technology in general to be empowering rather than coercive.
Comments
Posted in News | Leave a comment

Facebook Stored Hundreds of Millions of User Passwords in Plain Text for Years

Source: Hacker News

Article note: Even the biggest, highest profile web folks are absolute amateur-hour bullshit. Is it because all the web tooling is bullshit? Probably. Is it because they don't give a fuck? Also, yes. Is it because we build complex systems way beyond our ability to manage without thinking about the consequences? For sure.
Comments
Posted in News | Leave a comment

Matt Taibbi finally makes sense of the Pentagon’s trillions in off-books “budgetary irregularities”

Source: Boing Boing

Article note: The U.S. military-industrial complex truly is the world's most amazing display of the iron law of bureaucracy. The influence of that kind of lucre is so corrupting that the process of being selected and act of touching it to attempt an audit appears to instantly put an entity in on the grift. I absolutely believe it is possible-and-easy to over bean-count and get in the way of doing things more than you save, and also to under bean-count such that the grifting grows out of control. Somehow, via perverse incentive, the DOD has done both, and produced a system which spends an unreasonable amount of energy and money on obstructive bean-counting, without actually having any idea or control over the flow of resources.

The finances of the US armed forces have been in a state of near-continuous audit for decades and despite spending billions of dollars and thousands of person-years trying to make sense of what the military spends, we're no closer to an answer, and no one disputes that there are trillions of dollars' worth of unaccountable transactions (but importantly, not trillions of dollars in spending) that make it impossible to figure out whether and when and how the Pentagon is being ripped off, or wasting money, or both.

Enter Matt Taibbi (previously), who is one of journalism's princes of incandescent invective, a superb polemicist at short and longer lengths.

But that's only half of the Taibbi story: the other half is his uncanny knack for unravelling baroque scams, cutting through the mind-numbing complexity and getting right to the chase.

That's what he's done in The Pentagon’s Bottomless Money Pit, an 8,000-word explainer on the Pentagon's budget crisis that is one of the clearest pieces of financial writing I've ever read, drawing in the structural, economic, personal and historic elements that have created the "bottomless money pit" that is the US military.

The problem is a snarled knot of many smaller problems. For example, the Pentagon has terrible IT systems. Leaving a "quantity" field blank in a purchase order form caused the computer to place an order for 990,000 units, for a total of $3.5 trillion. The order never went' through, but the system also had no way to unwind the transaction, so it was just left on the books, and mysteriously deducted later, creating an accounting overhang a third the size of the US GDP.

When the military manages to actually order things, it doesn't keep track of things. Key military materiel like nuclear weapons have historically not even been assigned serial numbers or tracking tags (the Air Force once accidentally shipped nuclear nose-cones to Taiwan, where they were expecting a shipment of helicopter batteries) (oops).

To make things worse, the system is genuinely full of waste and pork, which any kind of real audit would uncover. The military contractors who benefit from these scams have gotten so rich from them that they can afford to buy key Congressmen on the relevant committees, and so every legislative attempt to force the military to genuinely account for itself has died.

Which is not to say that there haven't been audits. There have. These audits have run for years, cost billions (literally) and either concluded that the system was unauditable as it stood, and needed a complete overhaul; or have later been revealed to be fraudulent and had to be retracted.

And that's where those trillions have gone. The military hasn't lost trillions of dollars, but its books contain trillions of dollars' worth of transactions, only a small fraction of which are real, and the noise in the system lets grifter military contractors rip off the taxpayer for billions, and their campaign contributions have ensured that this will never be fixed.

The Pentagon just committed to giving billions more to Big Four auditing companies to conduct another audit, though the best we can hope for from all this is that they will simply repeat the conclusions of the other auditors who've gone before them.

The Defense Department, for the most part, does not know how much it spends. It has a handle on some things, like military pay, but in other places it’s clueless. None of its services — Navy, Air Force, Army, Marine Corps — use the same system to record transactions or monitor inventory. Each service has its own operations and management budget, its own payroll system, its own R&D budget and so on. It’s an empire of disconnected budgets, or “fiefdoms,” as one Senate staffer calls them.

Instead of using a single integrated financial accounting system that would maintain a global picture of its finances at all times, the Pentagon built another bureaucracy to pile atop the others, called the Defense Finance and Accounting Service, or DFAS. Created by then-Defense Secretary Dick Cheney in 1991, DFAS is in charge of collecting financial reports from all the different fiefdoms at the end of each month. DFAS is like a tribune traveling on horseback at month’s end, collecting a pile of scrolls from each castle.

In 2013, Reuters published a brutal exposé showing how DFAS accountants conducted a mad scramble at the end of each month to try to piece together records of transactions to justify spending. But in thousands of cases a month, no records existed. “We didn’t have the detail,” one accountant explained.

Complicating matters is the fact that money is allocated to the military on different schedules. If Congress gives the Navy $53 billion for operations and maintenance, as it did this year, the service is expected to spend all that money that year. Such expenses — payroll is another — are called “one-year money.” Meanwhile, research and development might be “two-year money,” and contracting might be “five-year money.”

The Pentagon’s Bottomless Money Pit [Matt Taibbi/Rolling Stone]

Posted in News | Leave a comment