splinter

I’m not totally sure what you mean by “get out more or less of the dissolved particles”, but I think I understand where your confusion lies. You keep referring to quantities, i.e. mass or particle counts. Their data is reporting these things as concentrations.

It should be obvious to you that 7.14g of salt dissolved in 2ml of water will produce a highly concentrated solution (saturated, in fact), whereas the same 7.14g dissolved in 350ml of water will produce a dilute solution. The concentration of the first one is 3.57g/ml, but the concentration of the second is 0.0204g/ml.

If somebody then turns around and says that 7.14g of salt dissolved in a mug of water will produce a concentration of 3.57g/ml, it should be readily apparent that they are incorrect. That is in effect what the authors are saying by reporting their results as particles/ml and then saying that those numbers are representative of what you might expect when brewing tea under normal conditions.

Does that all make sense?

They report their findings as particles/ml, not particles/teabag. It should be obvious to you, as a scientist, that the particles/ml evolved given 1 teabag in 350ml of water will be massively different from the particles evolved with 1 teabag per 2ml of water.

So have I, and I understand why they would have chosen this approach. My issue isn’t their bench technique per se, it’s in their calling equivalence to tea brewing at home and articulating conclusions based on that.

Your objection to my describing it as “blending” is fair. However, it would absolutely not be plain swirling. With such a low ratio of liquid to teabags the physical agitation will be quite significant. Most people do not have multiple teabags in their teapot all colliding with and abrading each other while steeping.

However, the biggest cause for retraction is their failure to report accurate volumetric ratios. They used 2ml water per teabag and then reported their findings as particles/ml. It should be immediately obvious that this cannot be equated to the particles/ml that would have been derived from using 350ml per teabag, and yet they never make that conversion. I’m not going to speculate as to whether this was a result of intent to mislead or a simple mistake, but it utterly obliterates their talking point of “billions of particles”.

Yeah, just don’t put your teabag in a blender.

I mean nothing about the methodology is even close to representing normal tea brewing behavior.

For starters, a typical cup of tea is around 300-350ml, not 2ml and certainly not 1, so the low end is already down to 23,371 particles even before accounting for the brewing technique.

Secondly, nobody holds their tea at an active boil while stirring it at 750 rpm. That’s virtually blending it. There isn’t a meaningful way to compare that to typical tea brewing behavior but I wouldn’t be surprised to learn that it produced 10,000x more particles.

No it doesn’t. This study is unscientific garbage and should be retracted.

Their “simulation” of making tea involved 300 teabags boiled in 600ml of water at 95 C while being stirred at 750rpm for an unspecified amount of time. They then took counts using undiluted samples of that liquid.

It isn’t clear why they chose such an absurd methodology, but it is absolutely spurious to draw conclusions from this about teabags used under normal conditions.

I don’t know that your comparison to Facebook holds water. Firstly, Meta’s employees are spread over three divisions: Apps, Platforms/Infrastructure, and Product Services (ads, strategy etc), where Facebook itself is just one part of the Apps division. Even assuming that Facebook occupies 50% of Meta’s total workforce (likely a massive overestimate), that brings us to around 30k employees for 3billion users, or 100k users per employee. That gives you about 0.5 FTE for your instance.

More importantly though, the job of administering a mastodon instance isn’t really comparable to the job of engineering a social network, so taking a Facebook’s salary or user numbers doesn’t really give us much actionable data. We don’t know how many Meta employees are directly involved in administration of Facebook, or how much they’re compensated.

Ultimately, it’s about what your users are willing to pay. If you can persuade all 10k of your MAUs that $9/month is worth the value they get from your instance, then go ahead. However, I suspect that you’ll be lucky to get even 1/10 of that.

The previous commenter makes a worthwhile point even if their phrasing isn’t to your liking. 8 people all making 120k per year at 32 hrs/wk seems excessive for a server with less than 10,000 monthly active users.

I don’t know man. It feels like pigeonholing somebody’s sexual preferences based on the style of their clothing might not be accurate.

Take a look at this photo of Mötley Crue from back in the day, and those guys were renowned for their heterosexuality.

It’s in quotes because the headline is quoting a source rather than reporting information that the newspaper has evaluated themselves.

I see what you’re getting at and your position is reasonable, but I think misses the point of the initial comment, viz. The Economist is known for objective reporting (neutrality in bias), in part because they are open about their editorial slant (non-neutrality of opinion).

For example: “Ukraine is winning the economic war. This is a good thing.” - Economist reporting vs. “Ukraine is winning the economic war. This is a bad thing.” - Converse-Economist vs. “Ukraine is losing the economic war.” - Pro-Russian bias

You made an assertion. If you are unable to provide supporting evidence, we can assume that your assertion is incorrect without needing to prove anything.