Re: Big problem there, BBC
Statistically the fact that they aren't used everywhere is just an anomaly.
256 publicly visible posts • joined 12 Apr 2006
I have had so many arguments over the years about why I've avoided GitHub actions. I can now say I told you so.
Admittedly most of the time the need to secure images, interact with on prem hardware was also a good reason. I did warn the free service would change to billed one day.
10 year of email addresses? I doubt I could remember all of mine. I certainly don't keep a record of them all.
What's the intent here? If they arrest you at a later date and spot a mistake on the form you filled din before arrival they have a case to lock you up?
I mean criminals will just break the law. It's only going to inconvenience people who choose to visit America. Fortunately there is plenty of the world out with the USA t visit. I do wonder how many academic conferences will migrate to Canada or elsewhere as people decide they just can't be bothered with all the hoops to jump through.
The plan is that by 2035 they will have a fully autonomous unit with AI to replace the man leading the horse. It will require line of sight to a custom series of base stations around the field and a perfectly flat area of land to operate on. It will have a guaranteed 99% uptime for the ai service and a wired internet connection to the machine will be recommended. Batteries will need to be recharged every 30 minutes.
I had a beautifully crafted c and assembly based application for a microcontroller which I was certain had no memory faults.
It had a memory issue when I reprogrammed it over the air. After weeks of testing I found the issue. It ended out being a bug in the avr GCC compiler.
My code was fine. No third party code was used. Apart from the compilation.
I should have just written it all in assembly...
Neovim isn't that bad. I still enjoy vim although no longer get to use it all day at work. I miss writing pure C and assembler in vim.
However, I've never managed to enjoy using Emacs. I still remember the look on the face of the PhD student that spent a week getting Emacs setup for programming VHDL for me and documenting how to use it and I had to point out I was using the mentor graphics tool chain as I needed a job after this project.
I have to admit he could output VHDL code fast. However, it usually took longer to work out the logic that the time saved wasn't really that worthwhile. Especially as I switch to verilog to save time and effort.
I have to agree. Why should I need to sign up to a third part service to find out government or MP specific information. Everything should be on freely accessible website with no tracking or login required.
Equally I'm fed up with every local club and organisation hosting everything on Facebook or x. I use neither. Update your website that has been abandoned for the last ten years and make information available to everyone.
All almost too. However, I really enjoyed the article and am too drunk to work out how to send a correction. Having spent the day despairing at pen test reports pre production I'm pleased to know that there really are people out there that don't think testing before production is normal. I thought not thinking about it before implementing was bad....
That's nice. I don't. If I can't tell someone what pre-existing condition or allergies I have and die as a result of that. That's okay.
Nobody will sue, nobody will be blamed but me. My family know my wishes.
If that slows down medical science I don't really care.
Nobody should have access to my medical history unless I want them to have access. I may change my mind at some future point. Again my choice.
Medical alert bracelet have existed for a long time. They tell the first responder enough not to kill the patient without revealing information to any random hospital/pharmacist/GP worker.
I don't want my siblings that work in the NHS viewing my records. I don't want my friends in the NHS viewing my records. I don't want any random person who gains access viewing my records.
Any records generated can go into cold storage and be deleted after a period I determine. If not, then I don't want you to record it.
Exactly. If they only allowed their apps they may have won. The reality is android is meant to be the antithesis of apple. Their actions and words worked against them.
I don't have an issywoth Google setting a price for access. However, if I have an alternative means of generation money I should not have to pay Google.
I'd say the same for apple, however, their normal customers do not notice the restrictions.
Unless you have some insane corporate policy stating you have to use windows. In every situation just develop in Linux to run on a Linux server. You free up most of the computer memory and resources, only need to run what is required, and testing on your local machine can match the deployment.
If you have to use windows, question why you are still working in a company that mandates you develop in one operating system to deploy in another?
Alternatively, just run everything in docker containers and deploy an equivalent setup as your production system? At least this complexity would be scripted, repeatable and deployable on your test and production systems. With the bonus that all Devs would be working in the same manner.
The sooner Microsoft concede and Windows becomes a Linux desktop environment with a wine alternative to support legacy systems the better.
A windows sub system for Linux to let us get over the short term hurdle of porting code to Linux would be more useful.
I used to not care what keyboard I used. A colleague was fed up watching me fix mistakes. He gave me a proper mechanical keyboard for a day. I ordered my own by the end of the day.
I moved to a new company. I took my keyboard and mouse into the office and refused to use the crappy keyboard they gave me. We were in a small office and everyone had to share desks. I wasn't in the office every day. My keyboard was never at my desk as someone would borrow it. Eventually we bought everyone a proper mechanical keyboard.
I spend around 8-10 hours a day at my desk. I predominantly use my keyboard. Spending £120 quid for a keyboard which has lasted me ten years so far is worth it. I know have two filco majest touch cherry brown keyboards. One in work one at home. I rarely make mistakes typing. Change my keyboard and I make mistakes.
The tactile feedback, both movement and sound makes it easy to know when I'm making a mistake.
It's also strong enough to batter an annoying sales/support person over the head with and show no sign of damage afterwards. This is a major plus for any keyboard.
The fact that so many people who work with computers all day every day are willing to put up with cheap barely functioning computers, keyboards and mice makes me despair.
I am not a Microsoft fan. I'd be happy to see an end to them. Other than for work requirements, and even then not often, I've barely used their products since the late 90's.
I do still find it difficult to argue against the idea that you sell to a competitor for less than you can sell yourself.
Ideally we should be encouraging Microsoft charging more for hosting anywhere other than azure to encourage everyone to move away from windows based systems.
I've argued since 1998 that long term Microsoft will become a Linux desktop environment as no publicly listed company can afford the staff cost for developing basic os functionality. The money is in support, tools and unfortunately belief in one service is better than another.
Ultimately Microsoft charging more will long term make them irrelevant. Alternatively, they will adapt to provide a service people want to pay for. There is still the fact they have the ability to force people to pay excessive amounts for existing services. Long term people will move on to something else.
I know that everyone has their own biases. However, in the UK at least they attempt to promote those that have the best understanding of the law and that apply it correctly. Not those that advocate for one party and tend to extreme interpretations to suit themselves.
Whilst in theory this could change things. It doesn't have to. Most large government contractors will still adhere to previous guidance on the basis that the next administration may change the rules.
There is also a reasonable chance that this might motivate Congress and the senate to actually create laws to cover this which may prevent future presidents stopping things by dictat.
I am totally bemused by the fact that Americans can elect a king for a few years that can issue pardons to anyone they want for any thing ahead of time. It also depressed me that their courts are filled with political appointees. I've know some senior judges in the UK that hold very different personal views to judgements they have made. However, to them they have to apply the law. Not the party opinion. It's insane but interesting viewing watching from a distance.
The correct response is for those involved to continue to publish without a federal mandate.once the first few attacks get through pressure to reinstate them will increase.
Admittedly there will be no financial rewards immediately. Long term that may well change. Especially as some public and private companies may be willing to fund this short term on the basis that any company that supports it and attempts to follow the advice will be better able to protect themselves on the future.
It's really only an issue if you've already been identified either by publishing publicly or from standard police/military investigations. If you are stupid enough not to hide you IP and use a mobile phone to register for a service then you will be identifiable to governments eventually. If you thought otherwise you are blissfully ignorant. Equally most secure systems and methods of hiding your actions are under attack from government agencies on a daily basis.
Want a secure system? Setup a server with SSH access establish keys write a message to a log that's deleted hourly on empherial storage. Want to improve it? Generate a one time pad and encrypt your data and delete it as soon as it is read. Only perform the encryption /decryption locally or on a machine created for that action. Hell, trash the server and spin up a new one after each exchange, perform the the decryption on a separate drive/server and erase the drive/servers between decryptions.
Ultimately you will still potentially get caught by increasing the number of people you are talking to or are stupid and keep copies or reuse keys or someone finds they key or most likely someone realises what you are up to and monitors your computer/room/actions.
The reality is, for the police the best thing that can happen is people believe the system they are using is secure. They publish all the information needed to capture them and you catch them in the act and not tell anyone. If you are doing something illegal and have money or knowledge you can build a secure system to communicate until the first person that knows the system is caught and confesses. If nobody keeps copies and no originals exist then you are safe. Although in the UK if they can prove you have a password you won't hand over you could still go to jail.
I mean it's not exactly rocket science to pick up the language. It's easy to use. There is plenty of code to start from as examples.
Equally, I'm sure there are some rust Devs who can write C code to get the bindings they want/need to support rust in the kernel.
I've not done any kernel development in nearly 20 years. However, in that time I've got married and had kids and have to work. I don't have time to work on kernel development. In 20 years time when I retire I'll probably be willing to return. I may even do so once my kids are older.
Sounds good to me. You do realise people can still use services not from American organisations? In fact, we'd survive quite happily after a short adjustment period.
I at home host and run all my own infrastructure and systems. I'd not notice if they all disappeared overnight.
However, no sane business is walking away from the EU without a legal order to do so and even then there would be an alternative service spun up quickly enough.
I've enabled text only email in outlook 360 on my account at work. It truly pisses off our marketing team. However, it has proven very helpful as my emails always reach suppliers. The idiots had enabled some tracking nonsense in the email signatures they tried to force everyone to use. Suppliers mail server rightly blocked them as spam.
I would happily agree to limited performance for all new drivers until they have driven say 50-100k miles. Unfortunately I am completely against tracking individuals usage of a vehicle. So 5 years seems reasonable.
I'd even accept 5 years or you can get a larger vehicle but only drive it if carrying children under 10 years of age. If you didn't learn to drive until you have kids trying to drive more than one in a small vehicle becomes comical.
No kids, no driving. It may encourage the teens to take young cousins or family members out to get a more powerful vehicle. I guarantee the familial response to the babies being injured will be more painful then the jail time they get. I'd even agree to only your children for the first five years in a more powerful vehicle.
I'd probably insist you were insured on a vehicle as a named driver for that period also.
I try to ensure that all web products my company develop only contains code we host ourselves. This means I have a cached copy of npmjs, nuget and other packages. It also means that these services fail I've a backup and can continue without issue.
I host fonts,scripts images etc for all sites on our servers. I have scripts setup to measure API performance and verify that there hasn't been a change over time.
All of these things can be done and lots of them can be automated. However, they require you to design them into the system and stick with it. It also generally means that you marketing team aren't going to get everything that they want.
Our company website however is the antithesis of this approach.
I switched our systems to postgres in 2008-2009. I'd previously been a big fan of mysql however even then could see postgres looked like a safer bet. The oracle takeover convinced me it was the correct move and all the fallout ever since has just reinforced my belief that I picked the correct tool for the long term.
I looked into this in around 2008 as the 5g standards were being finalised using a very early licence and hardware and it was a great feature. Guaranteeing voice or emergency service access and performance is one of the better use cases. It makes a lot of sense.
I think it is less likely in the current form to be used for applications fast lanes. Equally, I can see the attraction for those that can't wait 30ms longer for data to pay more for a faster connection or need to download/upload their 100 GB file as fast as possible. I think if reasonable minimum performance were to be guaranteed for everyone else fast lanes could be a valid way for mobile networks to fleece the terminally stupid.
Ultimately all mobile data is going to become data packets. Be that voice or normal TCP traffic. Personally I'd prefer my 999 call to be guaranteed a perfect connection over your random YouTube video. Equally I don't need my text messages to be received in 100ms. I can wait so shove them in the slow lane.
I see the issue more one of piss poor regulation than net neutrality being the best thing we can aim for. Guaranteeing better performance to someone who will pay more may or may not work for a company. However, the regulator should be ensuring that everyone else has an acceptable performance.
Why shouldn't a company or individual be able to choose to pay a lot more money to guarantee performance. We do it with fibre and other pretend broadband connections.
What we need is sane regulations for minimum performance for the majority that increase over time at a sane rate.
As this won't happen. Keep net neutrality and go after the bastards.
If all the advertisment did was show me something relevant and not track me. I could possibly accept tracking which site a click came from if it could not be matched to a purchase in any way. Or possibly if you could guarantee that it couldn't be tracked to a specific user. I don't necessarily mind the company I buy from knowing I've bought from them before. I don't want them knowing I clicked through from an advert on the guardian and letting that leak to the world. I've a reputation to maintain.
When I want to buy a lawnmower I'm happy for adverts showing lawnmowers. I'm not happy to still receive them three weeks after I bought one and am trying to buy a computer.
Have an advert for a SAP on the reg when I'm viewing stories about a disastrous oracle deployment is amusing, appropriate and may very well target a group of people that are interested and can purchase these services.
Having the same advert on YouTube if my kid is looking at some pointless video isn't useful to anyone.
You can ride a moped at 16. That is why you can apply from 15 and 9 months. To ensure it arrives on time. You can learn to drive off the public road network and sit your exam on your birthday. I tried to do this to get my licence before the new photo licence was introduced. Unfortunately my birthday was the month before the change and I couldn't get a test booked within a reasonable distance from my home.
https://www.gov.uk/ride-motorcycle-moped/bike-categories-ages-and-licence-requirements
My five year old can brew a beer and could drink as much as he wants. Licence laws only prevent the purchase of alcohol for those under 18. Unless they are 16-17 and accompanied by an adult and eating a meal.
https://www.gov.uk/alcohol-young-people-law
1.4 MiB is correct.
Fun with numbers in computing has been on going for years. Probably why mebibytes were standardised. It happened when I was starting uni around 1998 which led to all sorts of fun in exams and assignments with lecturers using MB to mean either MiB or MB especially when they realised that they had to accept both and update thier questions and marking schemes.
Mega bytes
1.44 * 1000 * 1000
Mebibytes is 1024.
I think someone made a mistake and used base 10 at some point and it stuck.
The correct size is determined by multiplying the number of tracks, sides,
sectors per track, and 512 bytes per sector, then subtracting the bytes required
to format the disk, and then dividing this figure by 1024. For a "1.44-MB"
3.5-inch floppy disk, there are
80 tracks
18 sectors per track
512 bytes per sector
2 sides
Multiplying the above gives you 1,474,560 bytes. This is the unformatted size.
To determine the number of bytes formatting requires, you need to know how many
bytes are used for the boot sector, file allocation table (FAT), and root
directory.
There is 1 sector used for the boot sector, which is 512 bytes; 18 sectors for
the two FATs (9 sectors each), which is 9216 bytes (512 * 18 = 9216); and 14
sectors for the root directory, which is 7168 bytes.
NOTE: There are two ways to arrive at the 7168 number:
224 entries * 32 bytes per entry = 7168 bytes
-or-
512 bytes per sector (14 * 512 = 7168 bytes)
Adding these figures gives you 16,896 bytes.
Subtracting the amount used for formatting from the total unformatted size gives
you 1,457,664. (1,474,560 - 16,896 = 1,457,664 bytes)
Dividing the above figure by 1024 bytes generates 1440. (1,474,560 / 1024 = 1440
KB)
To convert to megabytes, divide by 1024. (1440 KB / 1024 = 1.406 MB)
Hmm, this is the register. Are you aware that people troll for fun?
Also, who doesn't have backups that they test regularly. Fuck you bad guys I'll restore from backup lockdown your entry method and apologise to my customers with a month's free access and get on with life. If I can't recover in less than 3 hours I haven't designed the system correctly.
I expect my team to be able to recover in that time without any effort. Admittedly every time we try we discover something wrong on the documentation. Equally we manage it within the three hours.
A backup is only as useful as the procedure to restore it. It is only a valid backup when you test the restore process.
E.g. I remember a fun period during my master's course when both I.T. people responsible for our computers were due to go on holiday on the same week. A mistake that should never have allowed. The solution was to appoint two students to work in I.T. for a week to provide cover. Actually I think they started before they left and finished after they returned. Being ignorant of how things worked and terrified of the .I.T. professionals they went out of their way to test everything they did during that week. It was made clear that everything they did would be tested, verified and they would be held responsible for every mistake.
Holiday over, the I.T. team returned in trepidation. What had broken? What did they have to do to fix everything? The students had discovered the backup scripts pointed to a server that didn't exist and failed silently. They pointed it to a valid server and set email alerts on failure. They created scripts to recovrer systems and tested it. They upgraded all the Solaris servers to the latest versions which had been an issue until that point.
Then they installed the tape decks and verified the backups worked and documented the recovery process.
I.T. were so terrified that anyone would notice what the students had achieved that we got away with anything on the networkg that year. Network gaming was no longer banned and we had outstanding games against staff and students. The students, learned a very important lesson. Test your bloody scripts regularly, then get someone else to check your scripts regularly and be certain you can recover from disaster.
Simply put, don't pay up, pay ahead and test your backups and know you can recover from an attack. It isn't difficult, it is what you get taught as an apprentice, as a student and as a professional. Don't be a lazy arsehole.
UKRI have a budget of £7,904 million in 2022 to 2023, to £8,874 million in 2024 to 2025.
The costs here are for the project managers that monitor the actual research projects. I've previously contracted PMs that do this job a number of days per month. So I don't see why they should be classed as inside IR35. Most of these people will only be doing this part of the time.