Have you done new benchmarks since Cloudflare announced their latest round of performance improvements for Workers?
Just curious if this workload also saw some of the same improvements (on a quick read it seems like you could have been hitting the routing problem CF mentions)
Speaking of `JSON` functions that can have drastic performance differences, V8 blog[0] recently had a post about improving `JSON.stringify` performance when you don't pass a `replacer` function. Some of the most used functions with performance pitfalls that are easy to trip into.
My understanding the "three doses" is referring to the 5, 10, 15 mL amounts not 3 doses a day. Later on it says that the amount itself didn't seem to have as much of an effect as dose + time did.
Its interesting the contrast between HN and X [0] as far as the sentiment goes on this change. On HN seems mainly in favor of this change while on X everyone is mainly against it.
Personally, I'm glad I haven't built something on PS Hobby that I have to migrate but I do think its fair to charge for your product.
I don’t fall out with the change at all, but being given 1 month to migrate doesn’t seem well thought out, there will definitely be some avoidable negative sentiment related to that.
These are potential paying users not freeloaders to evict.
This is the only real problem I have with how they handled the situation. But it's a big problem.
Whether or not these folks on free plans are ever going to convert to paid, they trusted PlanetScale to serve as a critical building block for their project/business. I think the least they could do is ease the transition by offering them a reasonable amount of time to offboard.
I personally would never trust critical infra to a company that has ever abruptly terminated a product offering with only 1 month notice.
With news like these, I always take the lesson: Nothing free lasts forever. If you are not paying for the product, the product doesn't last for you. You will have to migrate or pay eventually. They don't owe you anything if you don't pay.
As a former student that competed in both National and State TSA events from 2006 to 2012 (and won multiple trophies across events including the "webmaster" competition) I'm not really surprised by this outcome. That doesn't mean that its the right outcome but just a result of how these competitions are run.
The judges are all volunteers that may or may not have a background in the actual event they are there to grade. So you get someone that just shows up, reads the rules for the event and then judges accordingly to those exact rules. Which in this case since the website is hosted by Github it automatically gets disqualified for not meeting the rules of the event. This is a pretty easy thing to fix which is either A) Just have a custom domain that hides the github hostname B) host it somewhere else. The students here seem to have taken the B approach and have it up on netlify now.
I certainly empathize with the students here putting hard work into a project only to get disqualified on a technicality, it is a hard lesson to learn and one around maximizing your results within the rules as long as you follow all the rules. Hopefully they will be successful in getting the rules updated moving forward to be more clear about the default github page templates vs just raw HTML hosting.
my understanding was that they were DQed for using github, not hosting there.
Which is just bananas. I can't think of a single decent piece of software that doesn't use Version Control, and 99.9999% of the software I know uses git for VC.
We need a new format. This institutional nonsense is for the birds.
How about reading and adhering to client project specs? That is a legitimate and important aspect of engineering. The spec said no github. Jekyll is also a legitimate tool for "decent software". In the professional realm, you will be using code generating tools, so the argument that 'X is commonly used for software development and thus the requirements are boneheaded' does not hold up,
Sure, if the spec uses it's terminology accurately and is unambiguous, but...
> Template engine websites, tools, and sites that
> generate HTML from text, markdown, or script files,
> such as Webs, Wix, Weebly, GitHub, Jekyll, and Replit,
> are NOT permitted.
Github is certainly not a template engine website.
One could make the argument that it falls on the engineer to discuss the spec with the client - but these are high school students, and when they did attempt to discuss the spec with someone they were given the bureaucratic runaround.
We should have bought this to our TSA advisor, but the judge was a different teacher at our school. The problem we had was that we had to reach out to find out that we were disqualified and the appeal period was already passed. It would have been so simple to move to netlify hosting as can be seen in the github repo. It all stems from the fact that we misinterpreted the rules and had no ability to appeal in the short period allotted.
> It all stems from the fact that we misinterpreted the rules and had no ability to appeal in the short period allotted.
Those running this have dropped the ball at various places here, that's clear. I was thinking maybe a preliminary step can be added to the contest to sanity check review a project (and thus allow for addressing cases like yours.)
But regardless, I hope this has turned out to be a fairly generous silver lining for you and your team. The site looks great. Next step, get investors and get those tourist up in the air. /g
Yeah, this is indeed one of those life lessons that is inevitable. I just feel like this is the kind of thing you should learn after you've had a chance to pursue your interests enough to turn them into skills.
Programming is a joy. Engineering is a discipline. Customer relations is an unfortunately necessary workaround to a set of dumb problems.
I had something similar with a TSA event in the late 90s. My buddy and I took second place in the nationals for Computer Construction. We asked "why didn't we get first place?" and the response was "because your inventory had serial numbers listed. No one does that."... which was jaw dropping, because both of us had receipts from recent purchases at computer shops were every serial was tracked. Near the end of the event we found out that the judge of the computer construction contest was a teacher at the same school as the winner. Our teacher did nothing to contest it as we had a flight to catch. -sigh-
It's funny to read this now given I was a mod on RuneLocus during what I would consider RSPS's heyday (2006-2012ish range). It felt like the scene died off a bit especially when Jagex finally brought back OSRS in 2013.
That said it doesn't surprise me that certain servers have continued to grow and thrive over the years.
Id be very curious to see player counts from then compared to know, but at least in the quality of servers they feel comparable and even better than osrs in some aspects!
It was noteworthy change for me as I was considering PS for a new project but was a bit weary of the read limit and unwittingly hitting it. Now with the bigger numbers I feel that I have less to worry about and could just dive in.
And since as you mentioned from the post its a popular complaint so I figured others might be interested in this change as well.
It isn't something we're going to get done in the near term but if you'd like to discuss, I'd be happy to chat about it. Feel free to email me at ben@vantage.sh
Just curious if this workload also saw some of the same improvements (on a quick read it seems like you could have been hitting the routing problem CF mentions)