I remember at a previous job we used agile for all our development. Including FPGA development (free nerd hat if you know what that stands for). Long and short of it is that FPGAs are a reprogrammable circuit that you can do digital logic on. They're brilliantly fast and flexible but take a bloody long time to code on. Someone thought that we could agile this, which is like saying you can steer an oil tanker with an outboard. I spent every standup saying "I'm doing the same thing as yesterday, writing testbenches and hardware description code. This hasn't changed".
One of our classes at university was to create the conductive layer for an uncommitted gate array. They actually burned the silicon for some of the class in the department fab lab.
Ultimate write once code...
That's something that I would have enjoyed at uni. Wouldn't want to be the guy who found they'd not defaulted a case statement or something after they'd burned it!
Predicted chances of a Trump victory down to 18% on 538 now. Dropping day by day.
It's been slowly but steadily dropping for over a week now. But remember, 18% is the same chance as rolling a six.
Sure. I won't believe Trump is beaten until Biden's inaugurated but the 538 trend is at odds with those who were saying Trump was getting a boost from the debate and/or catching Covid.
The first rule of computing is Don't Use Excel For Anything Serious. It's fine for producing trivial budgets and pretty graphs, but absolutely useless for anything serious. Quite apart from anything else, it's extremely difficult to debug or audit, exceptionally prone to cut-and-paste bugs which won't be obvious, and exceptionally vulnerable to bugs caused by minor changes. Nearly all large Excel spreadsheet systems are completely riddled with bugs.
I actually do use Excel for something very serious, those pretty datawrapper charts you see in PB thread headers.
So is there an actual official figure for number of cases over the past few days? Or are we just guesstimating?
The figures on the dashboard are up to date, the extra cases have all been added in at this point.
Except that pinned at the top is this rejoinder. The cases by publish date for 3 and 4 October include 15,841 additional cases with specimen dates between 25 September and 2 October — they are therefore artificially high for England and the UK.
That's consistent with what I've said.
Yes - the missed cases have been added as back dated data.
So for the days that they were added on, the reporting-day numbers are much higher than they should/would be. As in "today we added x cases to the data".
They are visible in the by specimen date data as well - assigned to the correct days.
This is by specimen date yesterday -
This is by specimen date 2 days ago -
Not quite increasing at the rate of the infamous "not a prediction", but rather concerning none the less.
I know there's lots of focus on the fucktacular way the data has been procured/transmitted/stroed but surely the bigger picture is the actual growth in cases ?
Absolutely. The main take home is that cases are actually still climbing rapidly rather than, as the incomplete data appeared to indicate, levelling off. The rest is just nit-picking.
My XSLX into CSV script theory starting to make more sense.
They were using Excel as a database, Excel isn't a database.
But that doesn't make any sense. It's literally impossible to do that and it sounds like a political source told a political lobby journalist they were doing this.
I'm not convinced that everything that's in the press about this is true. Anyone who has had more than 5 minutes of data experience knows that it's not possible to use Excel as a db, even MS know that which is why they have Access for consumer db use and SQL server for businesses.
The columns for case data makes no sense and neither do the file size limitations. I can't say for sure what did happen but I'm pretty sure the first of those definitely didn't.
Of course it's possible to use Excel as a DB - people do it informally all the time. In fact, the main use of Excel is probably for informal record keeping rather than for calculation. I, for example, keep spreadsheets of things like expenses or bugs to fix. Of course there are tools better suited to such tasks, but Excel is great for quick or temporary solutions.
If you wanted, you could also write scripts to query and write to an Excel spreadsheet just as you would a DB. Obviously this is not at all a good idea, but you could do it.
But there's a whole branch of computing built around querying databases, why wouldn't anyone just use that instead of loading an Excel file into a dataframe and then using the analytical functions in pandas to query the df, it's a completely crazy idea and it adds unnecessary steps to the process.
This is definitely something that strikes me as too stupid to be true, especially given that we know we PHE have a pretty good database which the API queries and the PowerBI visualisations run from.
I'm all for scrutinising the government and poor decision making, but I'm also far from convinced that our lobby journalists and their political sources actually have any kind of clue about what actually happened. I do wonder why there is no official statement about it, there's just a bunch of misinformation and speculation dressed up as fact at the moment (including mine!). The lack of comms from the government has been extremely disappointing throughout this whole period.
Mr. Above, I look forward to seeing what action Boris Johnson 2020 takes that Boris Johnson 2021 considers unacceptable.
I think the options are no deal or extend on almost current terms until 2024 (that date of course not cynically chosen to put Brexit at the centre of the next GE).
I'd love to see Johnson running on "Get Brexit Done" in 2024. The British public are not the most shrewd and sophisticated bunch of people on this planet but there are limits.
If a tradesman comes to your house to repair your fridge and has nothing but a hammer with him, it is obvious to everyone, even Boris, that he is unfit to be a fridge repairman, professional or amateur.
If a data scientist tries to compile the official government SARS-COV2 data and has nothing but Excel, it is obvious ....
Predicted chances of a Trump victory down to 18% on 538 now. Dropping day by day.
It's been slowly but steadily dropping for over a week now. But remember, 18% is the same chance as rolling a six.
Sure. I won't believe Trump is beaten until Biden's inaugurated but the 538 trend is at odds with those who were saying Trump was getting a boost from the debate and/or catching Covid.
Unfortunately there's not been any update to the YouGov MRP since it was released on the 27th. That had a tightening in Trump's favour over September which was at odds with other prediction models.
Utter cr*p from the Welsh Government. This is testing devolution beyond breaking point.
I cannot see any good reason why one set of UK authorities should be able to look at the same scientific evidence and come up with a completely different set of highly intrusive restrictions on daily life than does another based on the same evidence. What's worse is authorities that insist on applying even more restrictive rules on one group of people compared to other groups, the main difference being that the former does not have the right to ultimately vote the authorities out whereas the latter does.
A lot of the confusion over the myriad of lockdown rules could be avoided if there were a single set of rules across the UK, with a traffic light system of severity applying in different areas, applied in consultation with the relevant local authorities or regional governments. There should on the back of the pandemic be a review of the extent to which current devolution of powers should be reversed during a national emergency affecting the UK as a whole.
On your logic, we shouldn't ban incomers from, say, the United States, because they don't have the vote in the UK.
My logic is that I would like one set of differentiated rules to apply consistently within all parts of the same nation state that both you and I live in, because the scientific evidence that should govern the restrictions on our liberties is the same and the UK government is accountable to all of its citizens. I would not object in principle to the UK government bringing in special restrictions on English citizens travelling into Wales, if it were sufficiently convinced by the Welsh Assembly that there were particular circumstances in Wales that should require it to do so.
A top-flight Italian football match - Juventus v Napoli - descended into chaos on Sunday when Napoli failed to turn up in Turin because of coronavirus.
After two team members tested positive this week, Napoli say they were ordered not to travel by their local health authority in Naples, the ASL. However, Italy's Serie A football league refused to call the game off. Napoli now face an automatic 3-0 defeat.
--------
Man United and Liverpool should have done that and taken the 3-0 defeat
The first rule of computing is Don't Use Excel For Anything Serious. It's fine for producing trivial budgets and pretty graphs, but absolutely useless for anything serious. Quite apart from anything else, it's extremely difficult to debug or audit, exceptionally prone to cut-and-paste bugs which won't be obvious, and exceptionally vulnerable to bugs caused by minor changes. Nearly all large Excel spreadsheet systems are completely riddled with bugs.
Comments
This is definitely something that strikes me as too stupid to be true, especially given that we know we PHE have a pretty good database which the API queries and the PowerBI visualisations run from.
I'm all for scrutinising the government and poor decision making, but I'm also far from convinced that our lobby journalists and their political sources actually have any kind of clue about what actually happened. I do wonder why there is no official statement about it, there's just a bunch of misinformation and speculation dressed up as fact at the moment (including mine!). The lack of comms from the government has been extremely disappointing throughout this whole period.
If a data scientist tries to compile the official government SARS-COV2 data and has nothing but Excel, it is obvious ....
(*A comment my sartorial choices often elicited when I was still working.)
https://www.bbc.com/news/business-54387856
NEW THREAD
https://twitter.com/SevillaFC_ENG/status/1313101312020680705?s=20