alttag

There's a big difference between "done loading" and "appears to be done loading". Google is reportedly very cautious about the latter, and pulls some hijinks to appear faster than it is. Frankly, I'd be happier if more sites paid a similar level of attention.

That said, if your search page takes over a second, you're right, it might be an issue with your network.

Years ago I worked with a developer who had a different methodology: 2x, then bump the unit of measure. Thus, 1 day -> 2 weeks; 2 weeks -> 4 months. It's been remarkably accurate over the past couple of decades.

A nitpick: don't use the apostrophe in the phrase "in the 1960s". It's not a possessive in that sentence.

See, e.g., https://www.thepunctuationguide.com/apostrophe.html

Does this break links and lead to link rot? Other StackExchange sites don't seem to have the page rank of StackOverflow. It also makes it harder to restrict searches to a particular site when questions are spread across multiple sites inconsistently depending on what year they were asked.

I have a question I want to ask, but haven't. This is mostly because I feel that asking it properly, describing what I've tried, linking to (not) duplicates and explaining why they're not relevant, and linking to the docs to try and explain why I don't understand them, etc., etc, is a great deal of work. In other words, explain why the top 20 Google and SO results and workarounds aren't relevant to the answer I'm looking for. Doing it "properly" would take about 30 minutes. ... and thus I haven't asked.

I went with a workaround that I'm unsatisfied with because it was easier than asking a good question.

My biggest complain about SO is that with the expansion into other fields that questions that used to be welcome on SO (or get more visibility because of being on SO) are being shunted to SuperUser or ServerFault or other sites under the StackExchange umbrella. [1] I think the drive to keep SO purely about "programming" runs into issues similar to the "No True Scotsman" fallacy [2]. As a developer, sometimes I need answers about Amazon Web Services, Azure, Docker, or the Linux command line, as a consequence of programming, but those sorts of questions are, more modernly, marked off-topic for SO.

1: https://stackexchange.com/sites 2: https://en.wikipedia.org/wiki/No_true_Scotsman

I send unauthenticated email on port 25, every semester, in front of my students, as part of a discussion on internet application protocols. I can't use "God", because the addresses are validated, but I do send "from" the school's IT director. I even give them the commands to do it themselves (along with a strict talking to about how it's not truly anonymous because their network access is authenticated).

I've been able to do it at every university I've studied or worked at.

Possibly several things going on, of which O2 may be only one. Research also shows taking breaks and exposure to nature both have revitalizing effects.

> Medicaid is a loan.

Um, ... no. Or, at least not in my case.

Our son was diagnosed with cancer at the age of three. Medicaid served as secondary insurance, and thankfully so, as the bills from the first month exceeded $300,000. This went on for nine months.

It's now been eight years since he passed, with no mention from any agency about recovery.

Besides, it seems you left a few things out of your diatribe. From the medicaid.gov link:

    > For individuals age 55 or older, ... recovery of payments 
    > from the individual's estate for [specifically enumerated things, 
    > which does not appear to include doctor visits].

Also:

    > States may not recover from the estate of a deceased Medicaid 
    > enrollee who is survived by a spouse, child under age 21, or
    > blind or disabled child of any age. States are also required to 
    > establish procedures for waiving estate recovery when recovery 
    > would cause an undue hardship.

So, at best a fear-mongering half-truth. Maybe we can agree Medicaid is for people with no money. The government seems to have a procedure that attempts to collect money from an estate, in certain circumstances, after a patient dies, if there really was money available.

My experience (as someone who was elected to public education administration) suggests it's often more about logistics. For example, adjusting how many bus drivers (and busses) are needed in a district to serve elementary, middle and high schools. Staggering start times reduces the need, but drives one of these options to be earlier than the others.

High school sports complicates logistics later in the day.

It's not just a question of learning; it's also a question of cost and resources.

I have no evidence to rebut the conclusion—indeed it aligns with similar research on high school students.

However, a big chokepoint at my current school (major state university) is the availability of classrooms. Classes are already booked in the old (and recently expanded!) building and in the new graduate building for 12-plus hours each day.

There simply isn't enough space to put off scheduling courses until 11.

Besides, just because something is good for the average student doesn't mean it's good for all students. My classes certainly have a different flavor depending on whether they're in the early morning or late evening. Even if it's the same course, the differently timed sections have their own character and attract different sorts of students.

I agree with you, but was disturbed to discover a couple of weeks ago that Google does not. A search for "define electrocute" returns a definition that includes injury, not necessarily death.

  > injure or kill someone by electric shock.

One of the more interesting lines of thought on Ars was that the request was so badly implemented (without a judge's signature; could have a been a National Security Letter instead; other similar reasons), that it was almost as if the person implementing it was doing it because he was pressured to by higher-ups, not because of true desire to see the request succeed.

Either way, I'm glad to see the negative attention the request received and even more glad to see the request withdrawn.

I had similar issues, but opted to go the route of replacing the battery rather than the whole Roomba. It's worked well for me.

In addition, Amazon reviews point to some third-party batteries that supposedly last even longer than iRobot ones.

So is content editing, in many disciplines.

I believe copyediting and layout is paid, however.

Yes and no.

An adjunct typically has a PhD or a masters degree combined with real-world experience.

Public school only require a bachelor's degree, but technically, teaching public school in most states also requires a teaching credential (especially with the No Child Left Behind law's "highly qualified standard), which is not part of most content (that is, non-teaching) programs. However, many districts are purportedly willing to hire a masters/PhD candidate and then back-fill coursework necessary for a credential (e.g., curriculum design, classroom management) as well as guide toward content certifications. These additional courses and certification exams are paid by the teach out-of-pocket, of course.

So, as with many things, YMMV.

Doctoral students, interns, part-time employees, entrepreneurs, hobbyists (employed in other job), ...

There are many reasons for a salary that low.

I'll admit, I wasn't initially sure this was relevant to HN, but this is a participant (customer?) perspective of a YCombinator startup.

> what state the SSN-holder was actually born

No, it was the state where the SSN was issued. Not all children applied for an SSN at birth. Centralization of SS offices also altered this practice.

See, e.g., http://www.ssa.gov/history/ssn/geocard.html

Don't know that's it's necessary to put the spoiler here. The hints provide an alternative to hosting a file.

I'm a bit peeved that much of what is ROT13-ed in this thread isn't spoilers. I went through the trouble to decode it, and typically didn't find it worth the effort.

Concur. But WASD works.

Your insurance costs are higher when you have a longer commute.

Your oil needs to be changed more frequently.

Wear-and-tear on your car is almost certainly correlated with miles driven, meaning maintenance, repairs, and replacement occur more frequently.

Cars ownership is about more than just the cost of fuel (which is cheap in the U.S. compared to the rest of the world). Sure, maybe the IRS rate is a bit high, but it seems the intention is to compensate for the average cost of ownership per mile, not the marginal cost to drive a mile. Don't trivialize it by pretending to be ignorant.

I think you're right, and there's a lesson here.

I was recently evaluating a several options on github for a tool I needed. All options satisfied my basic requirements, so the choice came down to community/recency, and documentation.

I went with the project that was better documented.

Joke was on me, there had been several great feature updates, but without new documentation.

So, please, I implore OS committers, as much as it may be a pain, please require documentation updates when accepting patches!

I agree.

In addition, I hope the next "fix" will be to cull the comments. While there are occasionally some worthwhile edge cases documented in the comments, there is also a great deal of spectacularly bad design advice, which doesn't help PHP's reputation among language elitists as being for script kiddies and beginners.

I also enjoy its companion library, Paris, an Active Record implementation on top of Idiorm.

https://paris.readthedocs.org/en/latest/

http://j4mie.github.io/idiormandparis/

The only time I every read the print version was in Chicago, where a street hustler gave everyone in my party free copies, and then followed us asking for money.

I don't expect that's the most common use case of a free paper, but were I _The Onion_ I would find it irksome.

I study software processes for a living. Most of the organizations I've looked at are large enough and have mature enough processes that some of your list doesn't apply.

In a recent project, we sent a request to an industry luminary (one of the signers of the Agile Manifesto), on what metric(s) would be ideal for improving the quality output of agile teams. He responded back that the only metric he thought worth considering was the number of interruptions developers encountered daily.

The company we were studying at the time made two distinct changes: 1) re-working their automated build failure notification email (better recipient targeting, clearer error highlighting) resulting in faster fix commits. 2) An organizational push to trim their unit test suite (by about half) which decreased time needed for hardening and freed resources for new development.

The security researcher quoted in the article agree with you:

  > Arbor researcher Marc Eisenbarth said there's no evidence
  > Healthcare.gov has withstood any significant denial-of-service
  > attacks since going live last month. He also said the 
  > limited request rate, the lack of significant
  > distribution, and other features of the tool's underlying 
  > code made it unlikely that it could play a significant 
  > role in taking down the site.

Although, I will say that if the tool is grabbing the dependencies of each page (which in total comes to about a megabyte per page load), it might add up over time. Yeah, it's static, and yeah, it's not LOIC (it'd surprise me if LIOC weren't also being used against this site), but the site debs didn't do themselves any favors with the size and number of their static assets.

I had a previous job that was this way. Their informal mantra was "Do it now, scrape the blood off the walls later." In practice, this meant poorly communicated expectations, multiple conflicting reporting lines, non-specific project task ownership (everyone owned it, or someone else did and you were expected to as well), and best of all, no formal (or even informal) QA process.

No surprise then, that between the various forms of attrition (whether from firing, as managers spread blame and scapegoated, or quitting from those capable and prescient enough to see it coming) the company lost nearly half of its IT staff in a 12-month period, which included a major software release.