Some multimodal models may have a hidden captioning step that may take completion tokens, others work on a fully native representation, and some do both I think.
Why is pay off greater in cycling than other sports? Salary of the top riders? Compared to say NBA players, pro cyclist make relatively little. Tadej Pogacar (best and top paid cyclist) makes about $8M (euros) in salary per year. Steph Curry (highest paid) NBA player makes $55M (dollars) in salary per year.
Basketball isn’t as demanding physically as cycling. You need to be fit but not to the extreme degree cycling demands. I would expect doping to be most beneficial in sports where pure physicality is needed. Marathon, triathlon, track running.
You can reasonably assume that some NBA players are using PEDs. However, the effect is different. To be an NBA basketball player you need to have several attributes, such as height and hand-eye coordination, that cannot be affected by PEDs AFAIK. If basketbally are using PEDs, it is probably to recover faster, which means coming back from injury or training more. More training can lead to a higher level of skill, but it's a second order effect. It's not like cycling where, for example, EPO directly affects performance on the bike.
yes, but those epo-esque drugs aren't exactly trivial to use these days. the testing process makes the doping process much more difficult for drugs that have these direct performance benefits.
recovery help is where it's at these days i expect, in most sports.
have you seen the physiques and workloads that nba/nhl/mlb players are dealing with these days? these athletes have more incentive than cyclists to dope ($$$), and the testing in those sports is a joke.
there are obvious performance benefits for traditional endurance sports, but the testing infrastructure is pretty robust and the financial incentives are much less than those big team sports. it's harder to dope (and get away with it) and the financial pressure is less.
I totally believe that a lot of basketball/football/baseball players take something. But the effect won’t be as important as in cycling or marathon or 100 m sprint where you need pure physicality.
The effect doesn't really matter. If it gives you a 2% edge, and you don't take it, then you're 2% off the top. That may be the difference between having a career at all and thinking about what could have been at your desk job.
Sure, there's no drugs that will turn you into prime Messi. But there are drugs that will let Messi play like prime Messi for 90 minutes, 3 times a week, 48 weeks a year, which is incredibly valuable.
The "pay off" the commenter is talking about is the results in the sport, not the monetary gain. Cyclists are like the engines in an F1 car. Not saying there is no skill involved, but any skill differences are irrelevant if the other guy is putting out 100W more than you over 200km. So it really comes down to raw power to weight ratio.
That's not the same in basketball or most other sports. You can't just jump on gear, lift weights and suddenly become Michael Jordan. Plenty of people could beat Pogacar if they could use anything they could, though, just like manufacturers could build an F1 car that would dominate every race if they could circumvent the rules.
Because beside some skill needed in going fast during descends at 70-80-90km/h without dying (which is not easy but not extremely difficult either), a cyclist is basically an engine. Most other sports need physical fitness (speed, stamina, strength, endurance etc) AND coordination skills, and the latter is not easy to improve chemically.
I could agree with this. You do need some physical gifts as far as muscular endurance beyond the capacity of most but after that, its a very limited set of movements performed over and over again for hours. Plus a massive amount of will power and pain endurance. No amount of chemicals will turn even most gifted people into an NFL athlete.
Not money. It's highly specialized in what physically benefits it, so even a small doping on that specific physical attribute leads to significant advantage.
Bikers and their teams are known for removing as much weight as possible from their bikes. Would love to see the math for weight/power/time ratio for a motor like this. Would it be worth it considering you'd have to expend additional watts lugging it around all stage? My guess is probably not. Especially on a mountain stage which is where the tour is really won or lost.
There is a minimum weight requirement for bikes. I remember reading somewhere that they actually add ballast to some of them because they can be made so light.
Not any more - nowadays being aero is more important and that adds quite a bit of weight. And disc brake sets are also heavier than brake pads used to be.
Theoretically, the motor would be most useful on the climbs of the mountain stages. On the flats a couple of hundred grams don't matter, especially when most of the leaders are hanging back in the group anyway.
That said, bikes can already be made under UCI weight minimums of 6.8kg. Yet from what I've seen, most tour bikes are in the 7-7.5kg range.
The difference between the top 0.0000001% of humanity and second place is very, very small. Fractions of a watt. Adding just 10W would be game changing, and modern lipos and brushless motors add far, far more power than their weight penalty subtracts.
A 60Wh battery weighs about 300g. That stage is about 5 hours. 300g seems a pretty small price to pay for a 10W boost, especially if you achieve it by making the bike be under the limit and then adding motor+battery (switchable with a dummy weight, of course) to bring it to spec.
(The motor, of course, would probably weigh more - but it remains the case that you can build a bike that weighs under the minimum.)
I think a better thing to do would be to outlaw algorithmic feeds where monetization is via advertising. If subscription based that is fine. The incentive for sub based monetization is to keep you long enough to continue subscribing. For ads it is to keep you on as long as possible which trends towards divisive / fear / anger inducing content.
Makes me think a bit about how negative content engages more people. Is this the same with people who don't like change? Not liking change activates people more than people who do like change?
I like the look of Gleam over Elixir for sure. I’d love to see some example code showing Gleam-based LiveViews but I haven’t been able to find it anywhere. Is it possible? Anyone have some code to point me at?
I like the look of Elixir over Gleam for sure because:
• Elixir allows for more flexibility and faster prototyping.
• Elixir's ecosystem is superior and more mature.
• Elixir compiles to BEAM bytecode whereas Gleam compiles to Erlang which then compiles to BEAM bytecode.
• Elixir supports Lisp-style macros.
• Elixir excels at web development, data processing, and distributed systems.
• Elixir's OTP implementation is better.
Since Gleam doesn't support macros, my guess is that Gleam isn't going to be able to run Phoenix-like or LiveView-like frameworks, but I haven't really looked at it, so I could be wrong (edited because what I said wasn't clear, and probably still isn't).
I don't mean to knock Gleam, but it feels to me like a project that came about to add static typing to what Elixir already does, and while it has accomplished that, it has failed, so far, to live up to all the other great things Elixir and Erlang do. I know it's still a young language, so maybe someday they'll get there, but they're not there yet. After coding for 33+ years, I just don't find static typing to be all that compelling or important to me, so I would never choose Gleam over Elixir as things currently stand.
It does offer advantages, but also disadvantages. Lustre lets the programmer pick whichever technique is most suited to each component in their user interface.
