893 karma · joined December 11, 2020
I may as well install this to reaffirm that yes, this is an Apple product (in case you fail to see the prominent logo) and yes, I am better than everyone else for owning one (/sarc).
Where you do stand to make a difference is with more casual observers and people on the fence. A show of patience and respect bolsters a good argument better than perhaps even the argument itself.
What does tend to change people's minds is forming good relationships with people who hold differing opinions and their desire to make the relationship work. Logic and rationality are secondary considerations. Hopefully they will accept better conclusions for these reasons, but it's quite unlikely without adopting the kind of approach espoused by Dig1t.
Twenty plus years later and I've still never had more useful interview-related feedback and I'm still grateful he was willing to share that criticism. Now, having been on the flip side of the process quite a bit, I especially appreciate how hard it is to provide meaningful "negative" feedback.
But then they did maintenance and broke the entire feature. Reconfiguring everything from scratch didn't work. A key part where a Docker image is selected was replaced with a hard-coded value including a long system path (and employee name -- verified via LinkedIn).
Because of constant turn-over in account reps we couldn't get any help there. General support was of no use. We finally got acknowledgement of the issue when we got yet another new account rep, but all they did was push us towards paid support.
We exhaustively investigated the issue and it was clearly the case that nothing could be done on our end to resolve it. The entire underlying compute layer was busted.
Eventually they released a newer version of the feature which did work again, but at this point it has become impossible to justify the cost of the platform and we're 100% off.
Good luck to them, but from my experience the business fundamentals are misaligned and it's not a company I hope to ever work with again.
At best, you can only try to layer in contextual info like this as metadata during inference, akin to how other prompting layers exist.
Even then, what up-to-date information should present for every round-trip is a matter of opinion and use-case.
The last thing you would ever want to do is associate your domain name with gross, offensive content like this. The web is crawled all the time for snapshot data.
Additionally, you're more likely to cause your own (potential) users to stumble on this than anything else.
IMO, the best policy is almost always transparency. If you were to redirect users (and referrer-based redirects are a fragile thing), send them to a phishing/spam awareness page and explain that they most likely arrived from such a source.
This basically sums up my "cover story" nicely so I didn't have to admit to myself that it was more about scoring points for my own position.
Dressing up vanity as integrity is a dangerous thing.
In this case, I think vasco's take is correct: the sysadmin was indeed trustworthy enough to exercise this discretion in response to overzealous employee productivity rules without at all undermining his primary responsibilities.
The proxy was definitely in a place to essentially trivialize it's impact. I'm pretty sure thats why it was placed where it was, as opposed to make it harder to find. If that was the chief concern, disabling logging would have obviously been the first thing to happen.
In hindsight, it equates to kicking someone hard in the groin for laughs. It hurts my foot, them a lot, and it's just a poor way to learn how to earn "respect".
I was sure the right thing to do was to report it to the CEO (there wasn't anyone else higher in the hierarchy to consider) for certain chastisement and correction. I was commended and the sysadmin got a browbeating.
I definitely did right by the policies in place. However, not too long after I realized there was certainly a better way to handle it and my behavior had more to do with scoring points than doing the right thing.
This was also someone who went out of their way to mentor younger employees and help kindle a passion for this line of work. I was a major beneficiary of this, so I look back upon that incident with nothing but regret.
- mobile support works, but is poorly done - very heavy library to load - RAM intensive; a few charts w/ lots (e.g. ~100) data points spirals out of control (e.g. memory spikes from 60 MB to 700 MB). On mobile this basically guarantees that the tab crashes. - UI interactions feel laggy and lack responsiveness (e.g. panning is pure guesswork)
I thought the SVG vs canvas focus would be nice, but not at these costs.
Lots of great data points (and a little bit of snark)
The clarifications on `-u` and arrays are useful. I'm definitely used to assuming newer (... non-ancient?) versions of Bash are what is available.
- It's almost always preferable to put `-u` (nounset) in your shebang to cause using undeclared variables to be an error. The only exception I typically run across is expansion of arrays using "${arr[@]}" syntax -- if the array is empty, this is considered unbound.
- You can use `-n` (noexec) as a poor-man's attempt at a dry-run, as this will prevent execution of commands.
- Also handy is `-e` (errexit), but you must take care to observe that essentially, this only causes "naked" commands that fail to cause an exit. Personally, I prefer to avoid this and append `|| fail "..."` to commands liberally.
A quick summary:
> Geophysical and hydrological observatories in sealed boreholes provide a powerful tool to understand the hydrology of crustal formations, a means measure the hydrologic signal from changes in volumetric strain, and a stable site for high quality seismic and geodetic instrumentation.
This data is not just useful on its own, but also when correlated to other research. For example, 400+ years ago there was a large earthquake (https://en.wikipedia.org/wiki/1700_Cascadia_earthquake) off the US Pacific Northwest coast (and corresponding tsunami in Japan) that is referenced.I hadn't thought about assistive technologies in general, so this is useful to consider.
For example, consider the example of Demand Media (e.g. https://variety.com/2013/biz/news/epic-fail-the-rise-and-fal...).
They invested quite a lot in cheap content generation and for a while it served them quite well. I struggle to imagine a scenario where lowering the cost of polluting would have made this strategy more viable.
https://www.experimental-history.com/p/i-wanted-to-be-a-teac...
For example, I like to compare snapshots to a rolling greyscale 3s average image and extract contours around large diffs. This still yields lots of noise (moving shadows are the worst!), but I don't need realtime answers and have a relatively fixed domain. This let me add models like CLIP to search out specific known false positives (and known good positives) to improve detection quality.
A moving camera, obviously has no fixed frame of reference to detect motion. As such, your heuristics need to change, and they'll get considerably more complex (and computationally expensive). Honestly, I wouldn't even know where to start, and likely wouldn't even be based on OpenCV.
Thankfully, far less "let's test on people", but Dugway, Utah has quite a history of testing and mass animal slaughters from fallout clouds and such.
In particular, I noticed that EventSource goodies are used to broadcast currently playing songs on each station, as well as to provide anon read access to their IRC server via a simple vanilla javascript webapp.
WAY cleaner and easier than web sockets for this use case, and CDN friendly to boot!
The real life story is pretty sickening: an ~8 hour ordeal starting with the CDP saying "we think you have drugs" and forcibly sending an 18 year old girl to the hospital for an x-ray, in handcuffs, where she is instead invasively strip searched (with the CDP and hospital staff in audience). Nothing is found so they eventually let her go.
The court basically ruled that:
- The CDP did its job "correctly". They just need to think you did something wrong to detain you and force you to accept medical interventions.
- This was the wrong way to sue: the plaintiff must seek state-law tort action against the doctor for malpractice.
Most disturbing to me is considering just is how many people were party to whole affair, and how easily a routine, but ill-fated encounter with authority can go so wrong.I can only hope that additional legal remediation brought some kind of justice and relief. Maybe this is yet another example of how contorted US law practice can be.
I'm not holding my breath.
My extra two cents: generally refuse to answer "why" and the "how long" questions. My personal responses are "because I won the drinking game" and "not long enough". I find these to be more far more truthful and useful of answers anyway.
I agree with counter points showing how animation can add useful context, particularly in response to user interactions. This fails to address the crux of the issue greenbit brings up: the 'me Me ME!' dance. Yes, there are some places animation is indispensable, but the problem is that 95% of them (and their ilk) are giving the other 5% a bad name. The best way to enjoy and maintain focus is to not have to be constantly on-guard for hijack attempts.
Add in the fact that so many things exhibit this behavior (animations just tend to be the worst), multiply by how many sites/apps you spend time with. Even major, paid products can be problematic. With the web, ads, newsletter/don't-go/"privacy"/etc pop-ups are bad enough already.
I think we can take things a step/leap further: icon/logo/front-page take-overs and how aggressively those are pushed in front of user eyeballs. I believe (citation wanted!) that publishing control over the Google doodles is subject to a very strict approval process to curate a very specific, subtle "experience". It has to be the most valuable space of virtual real estate in the world, even despite limitations. It's certainly the most obnoxious to me.
So if there is to be general agreement on how things improve, I'd rather it be based on something broader than accessibility, but if accessibility wins some ground I'd appreciate it nonetheless.
Something is going to need to motivate you through the process of learning, failing, and growing in a new field. If your motivation is primarily based in getting away from what you are doing now then I suspect you have a difficult road ahead regardless of whether you change or not.
But if you have some kind of passion about what you'd like to be doing instead then you've got the most valuable compass and sail you can ask for.
> "Their inability to suss out all possible downsides of a set of EULAs..."
This is the part I quibble with, however. This assumes folks even read legal-speak. Privacy laws have undermined this by forcing more text and more "I see, I agree, go away" links/checkboxes/buttons in front of people. The problem isn't the EULAs, the problem is that no sane person can (or should) keep up with it.
Yes, sometimes people can be surprising. My wife has an awesome t-shirt from MailChimp from a not-so-hidden "if you read this, you get a free shirt!" clause she found in their legal stuff. But relative to the amount of cheap "consent" given, this is a huge outlier.
That said, I also have to give kudos to some of the lawyers I've worked with in the GDPR/CCPA/etc era. At least at the point of purchase, there was no better corporate consumer advocate of "boil it down to two or three crystal clear sentences and link to the details, which must cut to the chase" to be found.
1. Individual allocations with strict "rollover" policies.
2. Unlimited, but reasonable, timeoff policies.
3. Individual allocations with 0% carry-over: if you would lose it, it's forced upon you. Please don't make us drag it out.
So that said, I don't see how you have a prayer in the world of avoiding "drag it out" unless you have a management culture thats firmly based in what dougmwne suggests: lead by example and you won't have to fight the extremes (which are unlikely to give anyway) for the sake of the core quite so much.
It's also more compatible with highly-geographical laws around unused vacation payouts. If you think your organization is winning, by minimizing these kind of benefits... I think you're definitely losing and it's putting your most dedicated talent at long-term risk for short term gain.
And thanks for a charitable understanding of the "covid baby" idea: obviously "all babies right now are covid babies" so this is a comparison of "pre-covid" vs "now" (and that early on, that line was much blurrier).