IP address is one of a large number of bits of data used to fingerprint a device and user. Changing your IP address and even browser does very little to impact how you are uniquely fingerprinted. These include (for web browsering); User Agent, HTTP_ACCEPT Headers, Browser, Plugin Details, Time Zone Offset, Time Zone, Screen Size and Color Depth System Fonts, Are Cookies Enabled,Limited supercookie test, Hash of canvas fingerprint, Hash of WebGL fingerprint, WebGL Vendor & Renderer, DNT Header Enabled?,Language, Platform, Touch Support, Ad Blocker Used, AudioContext fingerprint, CPU Class, Hardware Concurrency, Device Memory (GB). Now i vaguely remember a plan by Apple to spoof this information to truely prevent fingerprinting but that is separate from the private relay concept. With large amounts of data points which can be queried from your browser fingerprinting is simply about probability that any one combination of the large number of data points is unique.... which it is likely to be.

So refreshing to see really interesting comment thread. Ironically Slashdot comment threads usually degenerate into ideological hatred. Here we have an ideology just about everyone hates, or at least the implementation of and we have great discussion. Annnnyway my turn to jump on the Agile implementation (hating) bandwagon. I have experienced agile many ways, Business Analyst, Scrum Master, Scrum Team Manager/Delivery Lead, PM etc. My intro was as a BA when i was told 'no need to gather any requirements up front, just do it the same sprint the developers start coding'. Took a year or so for that idiotic idea along with the fact Solution Architecture could be done on the fly to quietly die (somewhat). Managing software teams where senior managers were like 'That teams done more points they are awesome' and even worse inherenting backlogs from other teams they had pointed and people saying that was 1 point it should have taken my team 1 hour and i'm like THEIR 1 POINT IS LIKE A WHOLE DAY. To be honest the implementations I saw were all water-scrum-fall. Projects with fixed deadline and scope delivered 'Agily'. They still wanted time and cost estimates up front. Which is how the real world works; people want to know what they are going to get when for what cost. Going to them and saying ' i can't tell you what you'll get at a point in time but i will tell you I'll deliver something at regular cadences' doesn't cut cut it. Unless it's agile planning from top to bottom in the organization it's not agile. That all said I definitely prefer to deliver in a defined scrum cadence. Regular requirements refining, demoing and feedback, assuming the size of the work warrants it and people are pragmatic about MoSCoW. In my mind it is critical to do some high level analysis and design prior to commencing. Otherwise you are at risk of compromising multiple non functionals e.g. scalability and performance. Also documentation is needed. People often focus on the technical side. There is often a large business change component which needs planning and alignment to the technical component. You need process architecture, training material etc. And then support material for 1st, 2nd and 3rd line support. At the end of the day my philosophy was always do what works and get stuff done and i've yet to see a perfect methodology or ideology. You need to understand the context you are delivering in and select the best tools (methodologies) for the job.

My S7 which is 5 plus years old still gets security updates. It's a myth Android phones don't get updates after 3 years. They may not for the most part get full mobile os updates after a few years but all the ones I've had have continued to get security updates.

I live in NZ and completely false. There is zero covid community based transmission here. The only ones who have it are those flying in and being tested positive whilst in quarantine.

The decommission costs of both reactors and storage facilities is mind boggling. Sellafield in the UK alone is going to take £100 billion and 100 years! German nuclear reactors take 20 years to safely disassemble and again â100 billion euros and climbing has been set aside. Then there is the storage costs if not only the fuel but the contaminated power plant parts...

Honestly I roll my eyes everytime the 'x nuclear disaster caused minimal deaths' argument comes up.The whole 'only 1 person died of radiation poisoning / got cancer' a) is factually incorrect (a basic read of wikipedia article on 'Radiation effects from the Fukushima Daiichi nuclear disaster' would enlighten you) b) most importantly misses out the vast majority of the economic and societal damage when 50,000 households were uprooted and had their lives destroyed. Many committed suicide, and most of living had a severe quality of life deterioration but hey don't let basic facts get in the way of you love for all things nuclear..

Yup fully agree content acquisition testing vs cognitive assessment... the later is what should be focused on but sometimes more work to assess so people default to easy route especially with massive class sizes.

Most academics agree with you I'd say. Having researched a lot on this it's a) because of interia b) its easily scalable so when you have a class of several hundred students... c) certain certification bodies require it. Closed book tests often focus on content acquisition testing vs cognitive assessment. For 1st year student courses with massive class sizes I sympathize but other than that they have zero purpose.

I did mkt research on 30 or so of these tools and with students help trialled a number for a Uni I work with. I'd say the least concerning part is the flagging. 30% of tests maybe flagged as high or medium risk.. the AI component is useless for actually telling you who is cheating. The tools are a) primarily there as a deterrent for the 'cheat curious' b) best used in conjunction with intelligent exam design e.g. question banks, randomised variables and question order, local knowledge questions etc c) for non human powered ones, used post exam as evidence once lecturers do exam results analysis and see results from other embedded plagiarism tools e.g. Turnitin, similar multi choice answer patterns and most importantly the variability of student results throughout the year. For AI ones you should not and wouldn't have the time even to review the alert flags for a large class. Do the post exam analysis and if the result looks suspicious or the student is suspected for another reason then have a look. Students should not be stressed about AI software flagging being the measure of whether you cheat. They should be more concerned with whether the software actually works, is stable etc and could impact their ability to take their exam (on top of the issues of the general online exam issues e.g. internet stability). And privacy standards and track record of the provider. Yes i'm sure thereare examiners who are using the tools wrong i.e. they think based on the results of the tool alone the student was cheating..a bad workperson and their tools and all but 99% wouldn't be that dumb as accusing a student of cheating is a serious process and if you get it wrong..

It depends if its human monitored or AI monitored. If human they'll get warnings and then escalated to appropriate party who may shut down. Vast majority of time and if AI would be left to run....if they've cheated test will be invalidated post.

Sorry that makes no sense on so many levels. Firstly a number of countries have freedom of information, official information acts. Secondly there are loads of types of information if you ask you don't get given in the USA e.g. govt classified, other peoples personal information.

You didn't even read his post properly..to busy foaming at the mouth as soon as someone said something true and non positive about Tesla. He talked about cars not stopping where they should which you never even addressed just ranted about idiots not realizing it then required action to continue when the car had stopped.

Nuclear power is completely uncompetitive, solar and wind costs keep going down and so do battery and nuclear just keep going up. There are now solar/wind battery combo bids going in lower than fossil fuel. Westinghouse was bought to the point of bankruptcy trying to build two. Every single plant built in the western world is horribly over cost and delivery time and taxpayers and electricity buyers have to pay over the odds for expensive power. That's not even including the decommissioning costs which again is often dumped on the taxpayer. Sellafield in the UK is costing over $250,000,000,000 to decommission and rising. Then you have all the nuclear waste stuck in temporary holding facilities because no one wants it stored permanently i.e. Yucca mountain. And then ever while or so one blows up aka Fukushima, and hundreds of thousands of people lost their homes over 80,000 permanently....so yeah I see the bright future it has.

Why cos he spoke the truth about Trump?

I'm pretty sure John Bolton is going to have something to say about that unless Trumps white house lawyers try and suppress it