|
As a not really math familiar individuum, every time I read a math related article and find some matrix formulas I need a mnemonic to imagine what is described there (if they refer e.g. to a matrix element like Aij). In german this mnemonic is "_Z_eile _z_uerst, _Sp_alte _sp_äter" what means "Row first, Column later".
What kind of mnemonic do non German speaking people use?
And yes; no the answer "not german speaking" do not need a mnemonic is out of question 
It does not solve my Problem, but it answers my question
modified 19-Jan-21 21:04pm.
|
|
|
|
|
Back when I studied maths at university (1964 or thereabouts), we referred to row-major or column-major.
Row- and column-major order - Wikipedia[^]
Cheers,
Peter
Software rusts. Simon Stephenson, ca 1994. So does this signature. me, 2012
|
|
|
|
|
Thank you for this. But in math, is it not _always_ Row, Column? You may see I'm completely noob on this 
It does not solve my Problem, but it answers my question
modified 19-Jan-21 21:04pm.
|
|
|
|
|
Yes, I think the convention of row-first is well established. Of course the underlying mathematics doesn't care where you spray them round the page.
I remember ordering as being the natural counting sequence
11 12 13 ...
21 22 23 ...
... and it accords with the "across the row, down the column" summation in matrix multiplication. The mental image of two forefingers scanning like that is burned into my brain.
 If only mathematicians could become zero-based. So often in matrix maths there are subscript expressions like p+q-1
Cheers,
Peter
Software rusts. Simon Stephenson, ca 1994. So does this signature. me, 2012
|
|
|
|
|
Quote: If only mathematicians could become zero-based
I prefer for math the one base aproach which avoid lot of N-1. I'm that much used to it that I have meanwhile no problem to translate the math to zero based code 
It does not solve my Problem, but it answers my question
modified 19-Jan-21 21:04pm.
|
|
|
|
|
Peter_in_2780 wrote: If only mathematicians could become zero-based.
That causes a lot of problems with number lines that do not have a zero. E.g. years go directly from -1 to +1; there never was a year zero. The elevator in our office building goes directly from floor 1 to floor -1. (To ease our minds, we have declared that the building does have a floor zero, but it is virtual, and the elevator is not built to stop at virtual locations.)
My wish is that we go back to the days of Pascal. If your array indexes run from 1950 to 2050, that is perfectly fine! If they run from 0 to 99, or from 1 to 100 - both are perfectly fine. Just tell what you want, and the compiler takes care of offsetting the index you are using in your program so that the element at the lowest index runs from the base memory address of the array. Compilers are quite good at doing that sort of calculations - it is a pity that the C group of languages forces us to do this sort of offsetting in application code.
Some hardware helps with trivialities: I once worked on a machine providing an instruction called MIX3: It decremented the value in the accumulator by 1, then multiplied it by 3. This was in the Fortran days: Fortran arrays are 1-based, and this 16-bit mini, adressing 16 bit words rather than bytes, used a 48-bit (3 word) float format. So MIX3 converted a "logical" (1-based) Fortran array index into a word offset from the (0-based) memory base address.
Handling discontinous number series, such as years around the start of our "Common Era", will always require special handling, though.
|
|
|
|
|
Perhaps I should have been more explicit. "Practitioners of linear algebra...." i.e. those who deal in vectors, matrices, tensors and their ilk.
Yes, Pascal arrays were nice. A large C project I worked on, someone wrote a bunch of macros to try and emulate that. Very clunky; modern C++ could do a much better job, I'm sure.
Oh, and here, floor numbering includes an implicit zero. A local example goes 2 1 G B1 B2
A lot of (old) instruction sets included features to "help" compilers. One of my favourites was "back axle" BXLE in the IBM 360 and derivatives. Intended to be the bottom of a FORTRAN do-loop, it used 3 consecutive registers to hold the running value, increment and limit. Do the incrementing, and branch if still looping (less or equal to limit). None of the FORTRAN compilers I saw ever used it though!
Cheers,
Peter
Software rusts. Simon Stephenson, ca 1994. So does this signature. me, 2012
|
|
|
|
|
perhaps off topic, but I suddenly thought about this mnemonic from vector analysis
"curl of a curl equals to a gradient of divergence minus Laplacian"
|
|
|
|
|
That's not a mnemonic, that's a book 
It does not solve my Problem, but it answers my question
modified 19-Jan-21 21:04pm.
|
|
|
|
|
0x01AA wrote: What kind of mnemonic do non German speaking people use?
Row, row, row your boat
Gently down the stream
Columny, columny, columny, columny,
Math is but a dream…
(Just kidding)
But I do remember the meme for sin, cos, and tan math my 10th grade teacher taught us:
Oscar Had A Heap Of Apples and Sally Counted Them
O/H = S(in)
A/H = C(os)
O/A = T(an)
|
|
|
|
|
The one that was drummed into my skull was WW2 surplus: Old hands always help other airmen. I hadn't met counting Sally before.
Software rusts. Simon Stephenson, ca 1994. So does this signature. me, 2012
|
|
|
|
|
Not kidding:
Quote: Row, row, row your boat
Gently down the stream
Columny, columny, columny, columny,
Math is but a dream…
I like it and I'm sure it is a "brain burner" (1:1 german english translation) at least for me.
Thanks.
It does not solve my Problem, but it answers my question
modified 19-Jan-21 21:04pm.
|
|
|
|
|
Call it "Fortran" vs. "Pascal" ...
|
|
|
|
|
Me and the wife are a bit baffled, and trying to work out what is going on.
This morning, she got a missed call from me on her mobile (android). Problem is at the time stamp, my mobile was powered off, in my backpack, sat in my office on the platform, a room which has no mobile signal. She is in Cyprus, I am in Qatar (well actually halfway between Saudi and Iran in Qatari waters north east of Qatar.)
We do have sporadic mobile reception from Qatar mobile operator Ooredoo, but do not know where this is provided from. Whether it is a vessel/rig etc. with a mobile station. Certainly doubt it is reached from the mainland as it full strength signal on my device.
Could this be a spoofing? or has someone cloned my sim or IMEI or can there be some other explanation.
Has anyone any idea how we can determine what the actually elephant has gone on?
|
|
|
|
|
Do you suffer of sleeping problems and do any sleepwalking?
|
|
|
|
|
I was in my office awake, just finished the morning calls with the onshore office!
|
|
|
|
|
Contact your mobile carrier? They will have access to their DB.
"I have no idea what I did, but I'm taking full credit for it." - ThisOldTony
AntiTwitter: @DalekDave is now a follower!
|
|
|
|
|
It could be spoofing (more likely) or a cloned SIM. Regardless, my concern would be that someone knows that calls are regularly made from your number to your wife's number. They might try to convince her, for example, that they've kidnapped you, the proof being that the call is coming from your mobile, and that she needs to send a ransom payment to wherever.
|
|
|
|
|
DaveAuld wrote: We do have sporadic mobile reception from Qatar mobile operator Ooredoo, but do not know where this is provided from.
I guess since the cat has already left the bag you might as well know about SimJacker. The carrier Ooredoo is on the list of companies affected.
Natural Selection obviously will eliminate these vulnerable SIM cards.
|
|
|
|
|
|
Spoofing is really bad where I am. I've heard of people getting phone calls from their OWN phone number. A lot of them at the very least have very familiar area codes or prefix to try to get you to answer it.
|
|
|
|
|
Kschuler wrote: people getting phone calls from their OWN phone number
That happened to me!
"the debugger doesn't tell me anything because this code compiles just fine" - random QA comment
"Facebook is where you tell lies to your friends. Twitter is where you tell the truth to strangers." - chriselst
"I don't drink any more... then again, I don't drink any less." - Mike Mullikins uncle
|
|
|
|
|
Someone asked me if I could give any advice about database clustering with PostgreSQL preferably on a Windows platform
It turned out hard to find a solution that was not Linux based, but I found some and added them here: database-cluster-managers[^]
Now I wonder if anybody has ever used something like e.g. SymmetricDS and how it will perform with lots of SQL write and read operations.
|
|
|
|
|
We use Citus, but on Linux. Not sure if it's what you are looking for, but it's worth a look.
|
|
|
|
|
Thanks, that's a well known name, it was even acquired by Microsoft so maybe a Windows version is in the pipeline
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
