j_not_j

It's not about atomicity.

Co-Array Fortran is for multiple copies of the same program running on different data -- SPMD. Think MPI but much easier to write for.

You start a CAF program with mpirun. Multiple copies on the same machine or multiple machines in a cluster, your choice.

Embedded directives to process arrays concurrently is a different topic, and confusingly called OpenMP with no "I". Thus your multi-core CPU can process a numeric code more quickly.

You start a program using OpenMPI with it's own name --- ./a.out for example. You can control the number of threads used. Etc.

"where institutions can sign up to using a federal DNS service provided through the domain registrar..."

The domain registrar is CIRA, and has only one of twelve board members having a federal government affiliation. See cira.ca for the facts. Their Canadian Shield services uses data from Akamai, Mozilla, and CCCS.

It is not "federal".

Sigh. Another comment from someone's memory that takes only 2 minutes to fact-check and discover to be incorrect.

I really like his two rules. They are much simpler than the usual collection of process and product definitions that usually come up.

Of course, elaborating in further detail is essential.

Of course, the contrarian position includes the fact that true system engineering organizations are actually publishing houses, producing hundreds or thousands of formalized documents. The actual systems created are byproducts. /s

It seems the province of British Columbia cannot support UTF-8 in birth certificates.

The child's name is λugʷaləs K’ala’ask Shaw.

Beware of silent corruption in your disk arrays.

https://www.nsc.liu.se/lcsc2007/presentations/LCSC_2007-kele...

Disks didn't always write a block in the correct place. Things are better now, but still not perfect.

I would recommend that you read all your data all the time; this ensures RAID-1 or ZFS will correct the silent bitrot. Hopefully. A megabyte or two per second read rate is all you need.

See the pdf for more. There is a git archive at https://gitlab.com/fsprobe for a reader utility.

There are a lot of unstated or untested assumptions in the article.

There are many categories of drug users, not just "addicts" and "party" people.

Some people have chronic pain issues and are subject to medically-imposed restrictions on their consumption of lawful drugs -- to the point where they seek out illegal "drugs" hoping to get something to alleviate their pain.

Some people are self-medicating other issues than addiction -- like anxiety. Maybe not a good idea from your point of view and mine, but there they are.

(I put "drugs" in quotes because they aren't really -- they are powders and pills sold with fraudulent claims they might contain the substances desired by the customer. And, perhaps, not contain poisons or toxic overdoses.)

Few street drugs contain what is advertised -- the Vancouver market is stuck with variable mixes of meth, oxy, sugar, fentanyl, methadone, occasionally some actual heroin, cocaine, benzodiazepine, even carfentanil.

Steet buyers receive the same crap regardless of what they want. It's a serious game of russian roulette when ingesting this stuff. Nalaxone is a common upsell for street "drugs" -- buy your stuff and your opioid antagonist in case the stuff is poison. (Nalaxone is free in BC).

Harm Reduction is a difficult proposition when governments have decades-old policies and international treaties enforcing prohibition. Putting addicts on lawful drugs draws easy political fire ("subsidizing fiends"). And how does a pharmaceutical company even get close to trying to develop and market a "smokable" lawful opioid (because that is what the customers want)?

Over 10,000 dead people in British Columbia (population about 5 million) due to opioid overdoses. Since 2016.

https://news.gov.bc.ca/releases/2022PSSG0056-001250

Software Engineering Risk Analysis and Management by Charette. This unlocked the meaning of "risk" for mme and how it applies to engineering and systems.

VAX/VMS Internals and Data Structures by Kenah. Not that anyone cares anymore but this provides a distinctively different description of operating system internals. Very different from Linux. Explains what goes on behind system calls; process creation; file systems and so on.

Structured Programming by Djikstra. Up there with K&R.

System Engineering Management by Blanchard. This introduces you to system lifecycles of very very large systems (pre-Google/pre-Meta). And how to manage the people building them and manage the engineering. Relatively hard to read; the NASA System Engineering Handbook or matching European references are easier reads.

C/C++ Users Journal (magazine subscription). Seeing a wide variety of programming styles, approaches to solving problems, and approaches to describing the solutions.

Various SQL language guides. Knowing SQL helps organize my problem solving even if a "database" is not part of the solution. "The" classic declarative language.

Other books already mentioned here: Mythical Man-Month; Visual Display of Quantitative Information.

+1 for Enemy of the State.

Gene Hackman, Will Smith, and the incorrigible Jack Black.

The scene where Hackman's character wraps a cell phone in the foil from a potato chip bag is hilarious; almost true as well.

The concept sounds good, but you also need people to review the FUSE implementation as well as muxfs. And test, and so on.

One question: do you plan to implement "concatenation" of filesystems, so you can build one very large muxfs system (10s of terabytes) ?

"Canadian telecoms resiliency working group"

It is ineffective. Their remit includes "architecture" but the Rogers network architecture is a clear failure.

There was no political will to evaluate the results of such a group.

A lot of gaps not described in the letter:

Lack of independent management plane for core business assets. Driving to the office is not a management plane.

No external visibility of Rogers functionalities. The Kentik analysis of BGP and reachability suggests their tech staff had a lot of analysis to do before understanding of the problem.

No fallback plan in case of a misstep. No immediate detection of a misstep.

Four hour intervals in notifications (which implies lack of visibility into issues and absence of plans).

Secrecy regarding social and first-responder impacts (censorship of 911 call volumes for instance) suggests Rogers views these facts as existential risk, or at least of very high risk to their management.

Pall-ette. Palette.

Loose. Lose.

Programmers who can't spel are allways anoying.

Some way of automatically applying 'less' to shell commands when the output exceeds a limit (like 40 lines) but is otherwise invisible.

(There may be issues with automatically not doing this when piped.)

> An in-language way to do conditional compilation...

Nim has some of this. But of course, it is not C.

So they are on sale? Time to buy?

/s

echo hw.perfpolicy="auto" >> /etc/sysctl.conf

What's the big deal, there are often changes, sometimes inconvenient, sometimes breaking changes, requiring a configuration file change or software update. This is true of any operating system update or upgrade.

For Vancouver, try T-Net

https://www.bctechnology.com/jobs/search.cfm

I rechecked using the UK site at godaddy -- godaddy.com/en-gb

and couldn't reproduce it. The steps were:

1. pick a novel domain 2. register and account with name and throwaway email 3. click checkout 4. empty the cart 5. enter a new novel domain name but don't Enter or Search 6. wait two days -- observe no email 7. click search 8. wait two days -- observe no email.

So I conclude they don't do the keystroke logging any more, or at least not on the UK site.

Godaddy does keystroke logging (it's their web page, why not).

Without your actually clicking "buy" Godaddy knows your domain.

Try it! Using a throwaway email, fill in the Godaddy form with a novel domain name, email contact, etc, but don't click on any "next" or "buy" or even "search" buttons.

Sit back and wait. The spam will start to arrive. The spam will be web and SEO and logo and app developers.

They are "affiliates" and they pay Godaddy for details of new DNS domains. They don't pay much perhaps, but they are getting first-hand info of your DNS ideas.

That's where the front-runners get the info; domain-squatting is free so it's free money to them.

> a collection of requirements adds up to a design.

No it can't A collection of requirements implies many, many possible designs.

Change the team, you get a different design.

A good-quality requirement is _testable_, amongst other things. After the design and construction are complete(-ish), the team should have to show that each requirement is in fact met by comprehensive testing.

Along the way, each requirement should be mapped to elements of the design. So you can (on paper) verify the design meets requirements. And the designer(s) can be confident they are meeting the full set of requirements.

And for the nay-sayers that complain you can't ever adequately elicit the requirements out of the users (or customers), then your engineering process needs to address this risk. See many comments that refer to ways and means to achieve this mitigation.

See http://ocaml.github.io/ocamlunix/

RCMP? They have nothing to do with enforcement in the City of Ottawa (local cops), the brief protest at Quebec City (also local cops), two brief weekend protests in Vancouver (local cops), or the multi-day border closure at Windsor (local cops).

The RCMP do have enforcement duties at the Alberta border crossing and BC border crossing, both now open.

Judging RCMP for the inactions of a useless Ottawa city police force is inaccurate and misleading.

The OpenMP project provides GPU support for the #pragmas which parallelize both C and Fortran 90 codes on SMP machines.

(Note: not OpenMPI).

See https://www.openmp.org/blog/openmp-api-speeds-up-autonomous-... for a description of an example application of this.

Both GNU and Clang support OpenMP pragmas.

Trudeau waffled and avoided the press for two weeks.

Then Biden called. Something about factory closures.

Then Trudeau made a statement threatening the protesters and the cops moved in at the big border crossing.

So I conclude that Biden cares.

What I would like to see (and haven't found yet), is a bash or python shell that automatically does a "more" when a command exceeds some number of lines.

So many times I accidently fill up my scroll buffer with an unintended awk or grep mistake.

I wonder how the designers "knew" or hoped that the S/360 instruction set was achievable with that hardware/microcode design.

Was it sufficient to cover all the instructions? The floating point portion seems like a likely risk.

(Perhaps, more likely, the microcode drafts were co-written with hardware designs so they had less risk of a failure.)

Nah, I don't think so.

Google is a subcontractor. The owner of the website is using a service to track browser behaviour, and, perhaps erroneously, user identities. The service is famous, but not unique.

What the subcontractor does with the data is the responsibility of the website owner -- if they allow Google to abuse the data owner (that is, misuse the PII of people visiting the website), then the website owner should be held responsible.

And none of the responsible GoC employees (I hesitate to call them "managers") was fired as a result of their mistakes.

Actually, that guess is incorrect, it's got a native pthreads with SMP support. Just not great support (for example, OpenMP is not available).

There is ongoing work to avoid the big lock and much of the network stack, for instance, is unlocked.

I'm guessing much of the OpenBSD trailing performance on many benchmarks is due to both memory allocation and filesystem performance. malloc has a lot of security features out-of-the-box; turning them off (don't do that at home) might improve the speed.

But basically these are not benchmarks but conversation topics. A "real" benchmark runs your own code, not someone else's.

As an interesting topic, the LAME result went from 9.94 seconds (OpenBSD 6.4, 2018, Intel Core i9-7980XE) to 18.476 seconds (OpenBSD 7.0, 2021, Intel Core i9 10980XE). So many things changed over those 7 releases....