Posts by Graham Cobb 1466 publicly visible posts • joined 13 May 2009
1) It is pretty stupid to get rid of a very experienced group of people giving you the benefit of their advice. You can always choose to ignore it if you disagree, and you may learn something useful.
2) It is not acceptable to stir up hatred against a group of people who have been considering hard questions and giving advice, by request. You may disagree with their advice but it is not acceptable to put their safety at risk.
Why do the left care about Twitter so much
This is really not about politics. I don't care what political views are allowed or disallowed on Twitter. Musk has bought it and can decide, if he wishes, that the only tweets allowed are retweets of things he says. Or any other message he decides. It is his to do with as he wishes.
However, he is not free to abuse people. The Trust and Safety Council bear no responsibility for any decisions or actions made by the previous management. They have been providing advice, based on their considerable experience and expertise, on request. They made Twitter better, and they made it more successful by increasing its reach to the communities they came from. If Twitter no longer wants that, that is fine - it will turn into Truth Social and become a tiny echo chamber for a bunch of losers but if that is what the owner wants, that is fine.
The individuals who advised, and even those who took decisions and acted for, the previous management do not deserve personal abuse. They were doing their job, in they way they felt best. It was just a job. If you don't agree with their actions, that is fine - instruct them to act differently or even fire them. But it is not acceptable to abuse them. Nor is it acceptable to fire up unstable idiots who may attack them.
I know I'm old but I guess I am missing something? "cringe" has been around for a very long time. There is nothing down-with-the-kids about it.
Sure, it is mostly used as a verb, but as English is famous as the language where "every noun can be verbed" I assume it is not unreasonable for someone to use a verb as a noun - "full cringe" may appear unusual but the meaning is perfectly clear in context.
PLEASE someone, convince me I'm wrong. I'm open to hearing a decent argument (no mud flinging if you don't mind).
If you are genuinely interested in "hearing a decent argument", try reading one or two of Mike Masnick's articles. Although no fan of Musk, these do not include mud flinging and are genuine attempts to analyse what happened at Twitter.
Of course, Masnick has his own views, which you probably won't agree with, but he does have "decent arguments" without mud-flinging.
I'm sitting up and listening today - the press conference should be about to start. This is an important - in fact vital - step towards the things you are asking for.
Of course there is a massive amount of engineering to do, along, I assume, with a fair bit more pure science. But this would be the best news of progress for a long time.
I guess we'll agree to differ. To me, those CVEs are exactly the reason browser storage is the best option for many people: browser security bugs are taken very seriously and fixed very quickly. And I stick by my assertion that they expose a very, very small remote attack surface - certainly not one allowing access to millions of passwords from one data breach!
Personally I don't use browser storage - but I take my security much more seriously than my family or colleagues do. I do normally recommend it to those I know won't be taking the sorts of pain I do to protect my security. For most people, the alternative to browser password storage is not more secure but much less secure - it mostly involves using variants on one or two passwords with some small changes - find one of their passwords and you can easily break into half of their accounts with only a few attempts.
Actually, browser password storage isn't a particularly bad idea. Sure it has weaknesses (device can be stolen or hacked, ..) but they are targetted weaknesses. The crim has to be either physically close or deliberately targetting you. Unless you are a prominent person, that is your best protection. Random hacker targetting company databases is unlikely to get your passwords.
Personally I use Password Safe. Actually I use a database in Password Safe format with various different apps - there are many, for all platforms. Some apps are FOSS if that is what you want, some are tightly integrated with Android and IoS (for example with "keyboards" for transferring passwords into apps), others are little more than databases.
I handle the sharing separately - outside the apps. I normally keep the master version of the database on my home PC. Devices sync by a variety of mechanisms, some of which only work at home, others are protected by OTP access so I can access them from anywhere if I have my phone.
However, all that is quite complicated. I normally recommend friends and family to use one of the commercial password managers. Even if it can be broken into, it is going to be more safe overall than using a word document!
Exactly. The discussions Twitter removed were not for being "lies". They were for being "misinformation".
Medicine, even more than some other sciences, is not exact. And simple statements about life and death are almost always false. However, many people do not have the training or experience to make useful inferences themselves and need to trust people with that knowledge. In other words, experts.
The "misinformation" was not much about factually false statements (those are sort of obvious and not many people believed those). It was about (apparently deliberate) abuse of statistics and misleading statements to sow enough confusion and worry to have a measurable impact on the effectiveness of the country's response to Covid.
It was completely clear that masks, vaccination, self-isolation, and the other recommended actions had a massive, measurable, effect on reducing both infections and the seriousness of those infections. Sure, they were annoying. Sure, they damaged other things (particularly the economy). Sure, you could debate what levels and timings to set for all the controls - particularly in trading off between the economy, personal freedom, risks. And, of course, it all evolved over time as society changed, the virus changed, and our understanding changed.
But misinformation attempts to influence that discussion with deliberate lies and misinterpretations. As most people just do not have the relevant information to make a judgement on that, it is in the best interests of society to make sure that anything that disagrees with the commonly agreed expert opinion be heavily marked as such. Otherwise we can't have sensible debates.
Free speech means being able to put your name to your speech without having repercussions.
It most certainly does NOT!
What you say has repercussions and effects. Some people may hate you or shun you for it. Your spouse may leave you or your children disown you. Your boss may even fire you for it. Other people may laud you for it. You might even be able to change some people's views. None of that makes the speech not "free".
Free speech means saying what you want to is not against the law. That's all.
Not something I saw. My parents lived through the war although my father first served immediately after the war, in Palestine. Neither they, nor their siblings (all older than them) were at all anti-Europe or anti EU.
They were, however, very anti Japanese. Some had served in the Asian theatre, and even experienced prison camps, and my parents lived as expats in S.E. Asia in the '50s. Both of those experiences turned them quite anti-Japanese.
While I have no idea whether later military families were anti-Europe, none of the people I know who lived or served during WWII were anti-Europe or even anti-Germany. I think any such concerns had been well erased by the Cold War, with the need to see West Germany as the frontier of freedom.
Is there? Can you point to some evidence?
I am willing to accept your claim that free movement has contributed to increased abuse in some parts of continental Europe, if you say so.
But I see no reason that means there is less abuse in the UK after Brexit than there was. On the contrary, I suspect the more extreme actions needed to enter the country (it being illegal!) have meant that a higher proportion of entrants end up owing a lot of money/control to bad people. That combined with the dire economic state, caused partly by Brexit, lead me to assume that abuse has probably increased.
But if you have some numbers to indicate otherwise, please provide links.
Well spacecraft (especially transports and space stations - rather than major missions) often carry all sorts of experiments and devices, and most of the software in those is probably written by PhD students who only care about getting the results so probably isn't very secure or bug-free. So a separate bus for non-core systems seems like a reasonable idea.
And are you absolutely certain that every aircraft you fly on has complete, and effective, isolation between the usb ports in passenger-accessible areas (including galleys, etc) and critical networks?
IANAL (but I am guessing you aren't either :-) ). My understanding is that "fair use" has little to do with value or scale. It is the nature of the use, whether it is transforming it into something else or just reproducing it, etc.
Also, of course, "fair use" is a US legal concept - there is no such established principle in UK law.
Errrr... how does using an Apple key allow me to log in from my Unix workstation, my wife's Windows PC, my Android tablet, etc?
And how do I log in to sort things out when my iPhone has been stolen?
A key is a separate physical token which works with many devices and which I can carry around on my keyring. And I can easily have another one at home if I want to have a backup login.
I much prefer the security key to be a small, light, physically separate token like a key or a card.
This doesn't stop the phishing. It limits the impact by making sure it can only cause trouble while the person is still convinced.
Passwords mean that once you are phished you are vulnerable until you change the stolen password (and if you are stupid enough to reuse passwords all your other accounts are vulnerable until you change all of them).
That said, this isn't new. Yubikey, and their competitors, have been offering this for some time. Although presumably Microsoft are making it easy to use and, much more importantly, easy to integrate into the enterprise.
(Mostly) it doesn't work like that. It doesn't matter what certs you have installed: the point is that the certs (all of them) are local. They can only be used to sign the transaction that is in front of you on the screen. They can't be copied by someone and used to sign something else in the future, like passwords can.
It doesn't particularly help with the type of phishing emails which convince the CEO to send a panicky message to the accounts payment supervisor saying "pay this NOW!". But it helps with the ones which convince someone to log in to what they think is your AWS root account giving away the password.
Upvote for the systemd comment... it may be crap but it is the option the real world has agreed to use to startup modern Linux environments. Live with it.
However, I don't think Microsoft want to make Linux extinct. I have said it before, and I have not seen anything in the last 5 years to change my mind: Microsoft aren't interested in the OS business any more, and would be very happy to let that be Linux. They see two sources of money:
1) The tax on all new end-user PCs (aka Windows licence). They know this is gradually disappearing but they want to keep their hold over client PCs (not servers) as long as they can. This is the only reason they continue to invest in Windows at all and they know it won't be long before the world stops being willing to pay. They have already prepared for the end of this stream by adopting Chromium as the basis for their browser so that they can just ship that, running on Linux, instead of an OS once they decide to exit Windows altogether.
2) Enterprise and personal software. This is no longer a software business, of course, it is a services business. This is where their future money comes from. All pay-as-you-use. For consumer and business users - with whatever device or OS they are using. Increasingly consumers will use Apple or Android devices and enterprises will use Linux PCs - they want to make sure they all use Microsoft SaaS apps.
Yeah, well, "5G" was hyped primarily by the US operators who saw it (i) as a way to sell the punters a new phone, and (ii) a cheaper way to deliver high speed broadband without digging trenches (particularly to young people who demand high speed but don't seem to care if it isn't very reliable - I guess TikTok videos are short). Helped by the boondoggle that is US telecom subsidies - which seem to generate free money whenever a telco can convince a government there is some "new technology" which needs to be subsidised.
The real technology changes in 5G are not bandwidth/speed - as you say, that doesn't require a new G. The real changes are in the network architecture: mostly around moving core network elements towards the edge, running on smaller and more commodity hardware and/or virtualised on commercial cloud and potentially allowing lower latencies for traffic that can be physically localised. But that isn't sexy for consumers so the press has been about bandwidth.
No, we're not moving on to 5.5G already.
One vendor, desperate to sell gear to telcos, has hyped up the latest increment in their equipment. Even the analysts seem to have called them out on it - Huawei are obviously not buying enough analyst reports.
Just ignore it. Everyone else is.
I think he will last for the rest of this parliament. No Tory wants an election until the very last possible moment.
Electoral Calculus predicts (https://www.electoralcalculus.co.uk/prediction_main.html) that at current approval levels there would only be 48 Conservative MPs left at the next election!! Even if Sunak manages to improve that I don't see it becoming a Tory win in the time left remaining.
Back when I used to run a software team, my favourite line was "if it isn't tested, it doesn't work".
I mostly used it to stop architects telling us we had to implement some arcane option in the spec that, as far as anyone knew, no implementation had ever used. But I occasionally used it to tell my team to go back and find some way to create an automated test for their sexy new feature to make sure it will continue to work in the future.
Kids of people who are on a Chinese watch list seem like very likely tracking targets, whether they have their own phones or not. If the kids are young, it is likely their parents are nearby wherever they are. For example, on a family holiday that just happens to be in a foreign city where there happen to be other subversives at the same time.
it’s rather specifically about software-type jobs, where productivity is both difficult to assess and varies staggeringly widely
While the article is about software jobs, that isn't where productivity is hardest to assess.
For example, I did time in business development, and also in marketing - both in highly technical companies, as a highly technical engineer myself. In both cases, no one else could really tell if I was doing a good job or not.
I am confident I was doing a good job - and everyone seemed appreciative. But they really had no way to know.
I saw some very tough sales managers - they micro-managed their sales people, wanting to know where they were all the time and if it was more than 25% in the office or less than 75% with a customer (or more than 0% pretty much anywhere else!) they would get a roasting. Even the ones who exceeded quota were closely managed (and got a higher quota next year!). But I, as bus-dev and some sales-support, didn't work for the sales manager and didn't need to be in front of customers all the time. As long as my sales colleagues made it clear they found me useful no one really knew what I was doing. Sometimes I was with customers, some times I was researching trends of customer needs, some times I was writing proposals, some times I was writing pieces for analysts or marketing, etc.
Even more true when I was in marketing. As long as I produced the things they "expected" from marketing (decent slides, a few great one-liners of benefits, some things to belittle our competitors, a stream of white papers, articles, analyst reports, videos, etc for them to send to their customers) they had no idea. I didn't even have to produce leads (we were a highly technical company in a small market - our sales people knew all our customers well and knew what opportunities were arising)!
I could probably have done two jobs in both of those roles without anyone knowing. But why bother?
When you look at all the events that our planet has seen since it was created, then it suggests that we have been very lucky to still be here
Or... it suggests that life is bloody hard to kill off and disrupt.
Without more data points we can't say which is the case. My personal guess is the opposite of yours: intelligent life is common as muck but the universe is so enormous we are very unlikely to ever interact with another civilisation.
Probably the first time I have substantially agreed with a post from Phil O'.
I don't agree with the underlying strategy but we can at least agree that it really is ridiculous when our Cabinet can't see the obvious! Unfunded tax cuts, just when the world is going to hell in a handbasket and half the country won't be able to pay its energy bills. Increased borrowing, just when global inflation is shooting up.
We may disagree on the right economic strategy but at least the adults in the room could agree that now is not the time for an untested economic experiment.
You say that yet we are now expected to have a deeper recession after the U-turns. A pro growth policy is not a bad thing and is needed to work off the debt. Not presenting how the tax cuts would be funded and the U-turns were the problem.
The world is having a deeper recession. I don't blame Truss for that (Putin's behaviour driving massive energy price increases is the largest factor, but there are many other global causes as well). But I do blame her for not recognising it, not dealing with it, and going ahead with the most obviously stupid, inappropriate and ill-timed policies.
Unfunded tax cuts? Rejecting energy company windfall taxes? Government borrowing money to give to rich people? Trickle-down economics at a time of global economic crisis? The sooner the election comes the better - even Labour won't be quite this stupid.
Left and right still have useful meaning, although not in the US where there is only Right and More Right.
I also think that most people, including most Americans, hate the Israeli government (for its policies and actions) but do not hate Jews. Most people can distinguish governments from people - just like with Putin.
And, of course, all politicians hate individuality (after all, they all entered politics in order to tell other people what to do) and plurality of views (it is makes it much harder to get elected if the people like some things you do and not others, and the same for your opponents).
That may well be their intention, but this is not an acceptable way to do it.
They can certainly terminate someone's account if they don't like them, but they must return all their money.
It is time PayPal was subject to the same laws and norms as banks.
No, but a reasonable approach would be for:
i) a discount on fines if the company immediately admits fault and fixes it, or
ii) a stay of requirements until an appeal is heard - at which time the discount is no longer available, and affected employees would be entitled to complain in their own legal proceedings that the company was aware of the problems from the date of the original notice
iii) no further delays available - if the company wants to appeal further they have to do it after acting on the order.
Who, specifically, owns those servers?
It doesn't work like that. Many different people own those routers - and they agree to work together to make routing work. There is no body that can veto anything.
In addition, anyone who wants to (you, me, Google, Facebook, my government, your government, ...) can add routers and can choose to which routers they choose to send their traffic. Of course, they need to make sure that replies (from people they care about) can reach them back again.
What that means, in practice, is that some big players (facebook, etc) and some big countries (US, China, EU, etc) can control their traffic any way they want. And other players have to play by their rules if they want to participate.
Err... unwinding is for much more than memory deallocation!
If you are used to an exception-based error handling model, it is extremely powerful but it means taking many actions other than calling destructors during unwinding. Various external entities (objects within the system, and external systems outside it) will require being told to stop doing something, to release resources or cancel reservations. Many housekeeping and tracing systems have to be informed about the unwind for logging, debugging and understanding performance.
If you don't understand the power of, and need for, stack unwinding and its handlers then you should never be allowed to raise an exception in the first place!
Yes, but that does not necessarily mean 3 days a week.
I have almost always worked for foreign companies, and since the late nineties with my boss in another country. As technology improved, we got a lot of work done with videocalls. Latterly I took to going to head office for about 1 week a month - sometimes just me and about one trip in 3 with my colleagues from other countries all coming at the same time.
Some people didn't like it but it worked quite well in my case.
...mutter, mutter ...grumble... "a remote, southern island of the Inner Hebrides of Scotland."
Since when do you need to tell people where Islay is? Surely everyone drinks Scotch? :-)
And it's only remote by US standards. It's got two ferry routes and an airport!!
Now, if you really want remote Hebridean islands...
It's no more our fault that they aren't choosing IT than it's our fault that they are choosing to be primary school teachers.
Do you have any evidence for that claim? I think there is a lot of blame to go round. And the lack of engagement of girls at school with STEM subjects, particularly at higher levels, is largely society's fault.
I don't want to force anyone. I want to excite. That means: investing in great teachers, interesting opportunities (we had a science club that went on trips to local technology sites), science labs, even just pointers to exciting Youtube videos (much better than the Shell, etc films we occasionally saw at school).
At the start it was regarded as drudge-work and was closely related to other jobs where women were strongly represented such as bookkeeping, accounting machine operator, data entry clerk. My mother had been an accounting machine operator and toyed with moving into computing before deciding to become an accountant in the '70s.
It was like in "Hidden Figures" - computing was regarded as an unimaginative clerical task until, all of a sudden, it became an exciting new area of research! But most of the new programming jobs went to men, leaving many women doing data entry punching cards or stacking cards and changing tapes as operators.
Yes, probably. And, of course, there is the complication that the current share price is less about the underlying value of the company and more about taking bets on the outcome of the litigation and how the money will move afterwards!
As one of the biggest internet players, I suspect many Twitter shareholders are fairly risk-averse and may well sell soon as they don't want to price in those risky bets! But I am no stock market analyst (I have no Twitter shares!).
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
