Posts by Mario Becroft

The reporting on this has been confusing.

Initially it was reported that the pilot could not counteract malfunctioning MCAS.

When I posted about how absurd/negligent that appeared, I was downvoted to hell because apparently MCAS can be disabled by flipping a switch.

Boeing now says new MCAS it will ensure the pilot can always overrule MCAS using stick alone. So clearly there was room for improvement. But all the same...

Now this report says that the crew "tried to undo [MCAS trim] changes [33 times]." And that they were fully trained on the type. So what went wrong? I agree that in the scenario of flying a complex aircraft like the 737 MAX 8, there is the potential for information overload, and with the lack of positive warning of sensor/MCAS malfunction, it might take some effort to gain full situational awareness concerning aircraft flight systems.

But if the crew positively observed uncommanded trim changes on 33 separate occasions--how many possible causes could there be? How long was the checklist that would have identified possible AoA/MCAS failure, leading to disengaging MCAS? In the event of multiple uncommanded inputs for an unknown reason, would it not be recommended practice in general to disable non flight-critical automated control systems and regain manual control of the aircraft? Would not multiple uncommanded trim changes clue in the crew pretty much immediately? How many other systems are able to input trim changes in that way?

Certainly 7 minutes is not much time to figure all that out. But nonetheless, the whole scenario remains puzzling to me.

What's in the fine print?

How many 9's of uptime are you promised as a paying commercial MS customer? What is the recourse if they fail? This is basic due diligence you would do on any vendor. I wonder if MS etc. are perceived as "too big to fail," and the (arguable) convenience of SaaS leads organizations to move to services that simply have no assurance of service level.

There is a real point to QUIC though--my inferences only, haven't read the (draft) specs.

The big one is maintaining connection state while moving between networks. It's great when the PPP session to your mobile provider hands over cleanly from cell-to-cell, but although this can happen as others noted, it can by no means be relied upon (especially if dropping in and out of coverage) and certainly not when moving between wifi networks.

Now, in my mind this might all be best solved by IPv6 and appropriate routing protocols. But we don't have those deployed in a way an average (or even technically-savvy) person can even remotely use. So, pragmatically, onus of fixing this increasingly falls to the application layer.

The dream is that no matter where you are, networks via which you might be connected, or plain inadequacies in mobile infrastructure...your connections should retain any state and continue to work. If you suspend your laptop or shut down your phone radios, 10 hours later and in another timezone, the conections should remain established and working the instant you pick up your phone again. They should also have differential service, so e.g. your html, css and js must be received fully and in-order at the expense of possible retransmissions, the RTP traffic as you talk can trade packet loss for reduced latency Techniques not used before at this layer such as FEC could reduce the impact of retransmission.

Not to mention revamping TCP's strategies for handling packet loss to recover much more rapidly in a fast changing network environment.

Everybody wants this. Apple and Google have been hammering at the problem internally and implemented some proprietary solutions.

For UN*X users, the mosh project has implemented the same concepts for some years now. It is breezy. I can have a dozen ssh sessions on my laptop, suspend it, wake up in another country, unsuspend, and in no more than 1 to 2 seconds, all my ssh sessions continue uninterrupted, with any output while I was offline being immediately refreshed.

This is the end-game and it ever more compelling. I will be interested to learn more about QUIC and see how it performs in the real world.

I guess I made the right choice in going with Atlassian suite from the start. It also cost less and has more functionality.

Conspiracy theory central

I struggle to guess whether the majority of responses to this piece are serious.

Apply Occam's Razor. Putin is a KGB man who mourns the fall of the Soviet Union and prides himself on doing everything possible to bring it back. The Cold War per se may be over, but the modus operandi remains the same. If anything, the FSB is perhaps more free to act than the KGB ever was; protocols for these things existed, now it is a free-for-all.

Anyone could throw out conspiracy theories one after another. The Russians have a history of assassinating defectors, or anyone they don't particularly like; I think the body count since the fall of the USSR stands at about 16, these being only those high-profile cases we know about.

Talk about false-flag and other conspiracies all you like, but the simplest answer remains that Putin's Russia is sending a consistent and clear signal to anyone who crosses them: you are not safe anywhere. This is their entire purpose.

I hope the Reg readership is above being taken in by the well-documented Russian disinformation machine. To sow discord and doubt has always been the most effective propaganda tool. Rise above it, please.

Pale Moon

I just hope there is a large exodus to Pale Moon, which continues to support XUL, with associated UI customisation etc. With an influx of FF refugees, Pale Moon, which is already well-maintained, could gather a decent number of new (core and extension) developers and become the new FF, in the spirit of what FF used to be.

Onwards and upwards!

Fortran 90 "weird?" I don't think so.

Fortran is a very common language (particularly in research and supercomputing).

Referring to the original competition rules, the allowed languages list is highly biased towards Algol-derived, object oriented languages. The competition would have been more interesting if other languages were included, e.g.: Common Lisp, Haskell, Scala, Smalltalk, SQL etc.

Killing spreadsheets for fun and profit

Spreadsheets just need to die.

If you think you need one you probably should be instead using a database that actually stores your data in a well-specified, consistent schema with named tables and attributes (not meaningless columns names like A, B, C) and type safety. SQLite is ideally suited to small-scale ad-hoc data storage/analysis tasks.

Dissent

Allow me to present a dissenting opinion.

I've found my Android Wear watch very practical. Maybe the UI could be improved, but the existing Wear UI based on simple swipe actions is a damn good effort and works well. I actually don't see a need for major changes.

Main uses I get from the watch:

- Telling the time. Why do you think wristwatches were invented in the first place and took over the market from pocket watches? Our phones are now our pocket watches. Any time I leave my wristwatch at home nowadays I find myself glancing at my wrist... oh... dig phone out of pocket, turn on screen... yeah that sure is just as easy as glancing at the wrist. No benefit to a wristwatch here at all...

- Receiving messages. Recieve an SMS, glance at the wrist and there it is. The majority can be dismissed with a flick of the wrist. If a response is warranted it can be made by voice dictation in a fraction of the time it would take to pull the phone out of the pocket. And the voice dictation is good! Less error-prone than the auto-correcting on-screen phone keyboards my friends seem to use (based on experience of trying to interpret their messages).

- Sending messages. The number of times I've been stuck in traffic and fired off a quick SMS using my watch's voice dictation feature without having to use my hands (illegal here).

- Customisable watchfaces. I can add whatever data strikes my fancy to my watch, which I'd probably never be able to find in a traditional wristwatch. Examples: ISO week number, 12 and 24 hour time, UTC time, sunset and sunrise points indicated on the dial, current weather report... All while looking beautiful like a traditional wristwatch. Do I need to take an umbrella with me today? Just glance at my watch. Sounds silly until you try it. When working in project management, glancing at the watch to get the week number is so handy. Doubtless there is a mechanical watch somewhere out there with ISO week, but does it have all the other features? Likely not. Point is I can make my watch work how I need and want.

- Apps. Stuff I could do on a phone, but it's so much easier on the watch. Like whipping out a calendar when discussing meeting/operational dates. Even checking my bank balances to see if I need to move money around before a big purchase--takes two taps to the screen. Done before I could have even got my phone out of my pocket.

- Novel applications which could never be done with just a phone. Guess what one of the best features of my smart watch is? It acts as a viewfinder for the camera in my phone. This is invaluable when trying to see the cabling behind some networking gear poorly packed into a rack, for example. I was on-site with a colleague just the other day, and this feature blew his mind. Sure I could find some remote camera hardware to serve the purpose, or fiddle around with a mirror on a stick, but this feature is there *in the watch already on my wrist*.

I've only scratched the surface here of all the smartwatch features I use on a regular basis.

If you've never used a smart watch then don't be so quick to dismiss its usefulness. If you have and disagree, I respect your opinion. But for me it all comes down to convenience. Sure, I *could* do many of these things with a phone. But nowadays I hardly ever take my phone out of my pocket except to answer calls. Smart watches have reached the point where they are unobtrusive and highly functional. In my opinion they are right in the sweet spot.

As to the people moaning about having to recharge it every night... how many of you wear your conventional wristwatch all night? Or do you take it off and put it on your nightstand? Thought so. Well, you do the same with your smartwatch, and it stays charged. There are zero instances where lack of battery life has been an issue for me with either of my smartwatches.

Disclaimer: I haven't used the Apple watch. My experience is only with the Moto 360 and G Watch R, both of which run Android Wear.

Mainframe and Fortran > LAMP

So what are they going to replace it with? Some unweildy Java EE platform that nobody really understands? Maybe MySQL and PHP/Python on commodity x86 hardware that throws runtime errors whenever the version of the PHP interpreter is updated? You have to be joking.

I feel far more secure with half the world's nukes controlled by a straight-forward, time-tested IBM mainframe designed from the ground up for reliability, using a language like Fortran that has been around for a long time, is not going away, and does the job it was designed to do well. BTW, Fortran is arguably the most popular language in scientific computing.

Replacing a bespoke COBOL HR program running on zSeries just because it is only 8 years old? Insanity. I bet the system they have works fine and does its intended job. If there is a problem with it, address the problem.

If this practice were applied in any other engineering discipline, we would laugh. What, we're going to tear down this skyscraper because it is 8 years old and there are now new materials that are arguably better? Replace that multi-million dollar bridge because it is a suspension design from the 50s, and these days concrete/steel truss is in vogue?

Cautiously optimistic

Haven't looked into the implementation, but I like the concept. There is no phone on the market today that has the combination of features I want. Being able to put together a phone that suits my very particular needs sounds fantastic in theory.

I am worried that in practice this won't really be very modular. After all the SoC and the screen are a large part of a smartphone, and comments seem to imply these are *not* modular. Talk about shooting yourself in the foot.

союз v. Ariane

I must say I am also curious about the economics of setting up an entire союз launch infrastructure. Why are they not simply using Ariane launchers--are союз and Ariane suited for different mission profiles?

Or did Russia lease the facility and set up the infrastructure themselves, competing on cost with alternatives?