Mux is definitely on our shortlist, the website and developer documentation made a great first impression. Very clear and easy to understand, the developer focus is obvious.
Really excited by all the incredible things are customers are building with video, and proud of our team. (Speaking of which, ahem: https://mux.com/jobs/)
Hey! Jon from Mux here. Curious about this comment:
> the business model won't tolerate the amount of video on someone else's service
Does that mean you aren't using S3/EC2 or the like, or is there something about how we've built our cloud platform that doesn't work for your business model? We've designed Mux to be a low-level primitive for video, like Twilio is for SMS, so I'd be interested if we're doing something that makes this harder for you.
Hey Jon! I've looked at Mux (mainly the careers page), and it's a great platform. It would be a great fit technologically, but I'm not sure that my business model (which is tentative admittedly) would cover infra costs for processing, hosting and consuming the amount of video I'm eventually expecting, as I'm running on a shoestring budget at the moment.
Plus, it's a good chance to for me to learn the ins and outs of video. It's not reflective of the quality of your platform, just a choice I've made early in the piece for curiosity's sake.
Mux supports both VOD and Live video (and seamless transition from Live to VOD); this service is only live streaming. We’re a crack team of video experts who work really closely with our customers. Better developer experience, documentation, and support. More feature-rich and powerful. 100% focused on building the best video products for developers, rather than being AWS service #213.
In general, we don’t like to bad-mouth competitors and we have lots of friends in engineering at Twitch and AWS (hi folks!), so hopefully this won’t be taken the wrong way. Twitch has built a great platform and kudos to them for this service, and obviously AWS is the most successful software business in the world today. Online video is growing quickly and there is room for a lot of players to succeed.
I'll go on the record asking a question that might be dense: why would this apply to stock buybacks and not dividends?
Stock buybacks accomplish a similar goal to dividends. You're transferring profits to shareholders. By paying dividends, you distribute profits via cash. With a buyback, you distribute profits by increasing the value of equity. There are tradeoffs between these two approaches (tax treatment and otherwise), but they do the same thing.
And aren't dividends (or future dividends) the ultimate point of equity?
> By paying dividends, you distribute profits via cash.
Dividends are paid from profits. This particular example of stock buyback would be paid from an emergency loan granted to the company under the assumption that it is to be used to help the company get back on it's feet.
Wasting an emergency loan on buybacks is the equivalent on burning the emergency cash by funneling it straight into the pockets of the company owners at the tax payers' expense, in a way that's a whole lot like fraud.
Companies used to take their profits and either invest in R&D/Capital if they were trying to grow, or give dividends if they were mature. But once C-level compensation became focused on stock options, top management have an incentive to increase the stock price over other concerns. If a company spends every dollar of profit (or cheap loans) on buybacks, they aren't spending money on things that will sustain the business long term like R&D, having funds to weather a downturn, etc. Not that every buyback is a problem. Apple has done buybacks because they felt there stock was the best place to invest their money. It's not like they blew through their cash account just to boost the price.
Because to earn dividends you must be a owner i.e. actually hold company stock.
To profit from buybacks you only need the option of buying the stock at a lower price than the buyback price.
The CEOs and other CXXs of these firms often have large amounts of options included in their comp packages.
This is why buybacks happen at market highs instead of market lows as sound management principles and common sense would suggest.
Increasing shareholder value by increasing equity value is just a pretext. The real purpose of buybacks is a swindle to funnel company money directly into the pockets of the C-Suite through what in the books appears a routine management operation.
These executives decide on the buybacks and exercise their share options just before, thus pocketing millions in company money and actually hurting shareholders.
Shareholders get the blame despite having little to no power and seeing their investments ruined and looted. Also, in many firms, a lot of shareholders are also employees.
Some companies pay out “retained dividends” on unvested RSUs.
Like buybacks vs regular dividends, the main differences have to do with their taxability.
From what I can tell, the main complaint in the Atlantic article is that CEOs time buyback announcements to coincide with their stock sales. The first example they give is the Home Depot announcing a buy back (apparently in the Feb ‘18 earnings call), and then selling stock after the insider lock up window opened. In all likelihood, that sale was scheduled well ahead of time (execs have to do this to avoid insider trading charges).
So, the controversy seems to reduce to Home Depot having a strong quarter and buying back stock, making money for shareholders, including the CEO.
Note that if they’d issued dividends, and they pay retained dividends on unvested RSUs, the net effect would be exactly the same (except taxes).
The controversy is not insider trading. It is executives exercising their share options during buybacks the price of which they themselves decide.
They decide when to buyback and choose to do it at market highs. This is not sound management and is not in the interest of the shareholders.
I learned about it in a recent edition of "The Intelligent Investor".
Most executives are shareholders only during the brief moment it takes exercising their options. They are not shareholders in the investor or even trader sense.
The claim that this amounts to distributing dividends is also not true. Options which is most of what the executives are holding don't earn dividends.
Because somehow it was decided that buybacks are greedy and dividends are good. Probably because your grandparents talked about retiring on dividends. But paying out the money in a different makes it evil.
Great umanwizard, it's a good thing you're here to tell us that Cambridge university is wrong! After all, what do they know about finance and economics?
This is clearly false: there are plenty of value companies with high volatility of profits and low volatility of dividends. Certainly I have seen loss making companies still pay dividends.
No, the definition is a payment from a company to stockholders that's labeled "dividend". I know of companies that were losing money that still paid dividends.
I am out of my depth here but isn't that because of a shortfall verses expected profits, or restrictions on moving money around the world (apple) verses borrowing huge amounts of money at low interest rates to give investors a tax advantage and giving management large bonuses?
Yes - this discussion is kind of ridiculous. If San Francisco had announced that they were removing elective CS from the High School curriculum, Hacker News would be saying "How can anyone expect to succeed in a STEM world without High School CS?"
I was one of the advanced math kids and took calculus in 11th grade. Then I went on to Calc 2 in college, and it was a different league entirely. I would have been better off building a stronger foundation throughout High School and then doing Calc 1 in college.
...which is exactly what the new curriculum aims to do. I have a 6th grade daughter in SF, and the strength of the new curriculum is that they spend a lot more time making math more intuitive. Instead of learning one approach to long division, they learn multiple approaches. In theory at least, this pays off in the long run for a lot of kids.
What American schools need isn't more acceleration (Algebra at age 14 instead of 15) - it's a better understanding of what mathematics actually is and why it matters.
What American schools need isn't more acceleration (Algebra at age 14 instead of 15) - it's a better understanding of what mathematics actually is and why it matters.
While I agree with this sentiment, my problem is with the one-size-fits-all approach taken in San Fran. Lumping gifted children in a class with everybody else does them a huge disservice...
My personal experience in 7th grade (pre-algebra) was horrible. My school decided to experiment by placing gifted students in math with the rest of the kids, with the idea being that we'd magically bring up the average performance. Instead what happened was the nerds sat in the back bored out of our minds and lost a year of math education (and this was with an extra teaching aide in the class - the two teachers simply couldn't keep the non-gifted kids on track AND provide us any extra attention). This left us all behind when we entered Algebra in 8th grade.
I hate to be "that parent" - but gifted kids have different needs than normal students and deserve the opportunity to excel without waiting around for everybody else to figure out 2+2.
Edit - I don't care if the advanced math offering in 8th grade is called Algebra or something else, as long as there is an advanced offering. The linked article made it sound like there was not such a class.
I think a solution could be to have classes based on age groups..giving kids 5 years to finish 5 years worth of curriculum so they can finish at their own pace and then grade them at the end of the five year term.
This way, if something is easy or interesting...or tough or boring, kids can choose how they want to tackle it. Having many different ages in the same class without the pressure to finish everything crammed within one year should help. Here is where maybe bright kids can teach kids that need help..or older kids can help younger ones.
Example: ages 5-10 study together in huge single room schoolhouses ..each one on whatever they want to learn. Ages 11-15 is another group etc. amongst them, they can be in sub groups according to interest or ability. A class can have 4-5 teachers who can tackle all of the subjects. Volunteer parents.
Test and grade them at the end of five years. Only test them every year or semester.
They should eventually be able to form groups...and work as peers. Is there really a difference between a 12y/o and a 13y/o...as adults, we don’t always consort with those born within the same solar year. Why wouldn’t it work for kids?
> Is there really a difference between a 12y/o and a 13y/o
There can be huge differences in social development there, and not even strictly tied to age. The variance is rather large.
But even if you just look at averages, as far as I can tell the average 13 y/o girl and the average 12 y/o boy are quite far apart in terms of their social interactions, starting with basics like "are you thinking about dating yet?"
All that said, getting 12- and 13- year olds to work as peers is a much simpler problem than the originally posed one of getting 5- and 10- year olds to work as peers. In _that_ context there are a bunch of problems, ranging from difference in attention spans to the basic issue that most 5-year-olds don't know how to read yet, and most 10-year-olds do, and so conveying information to both together in a way that's not frustrating to one or the other can be quite difficulty.
Now if we want to group students by ability (5-year-old who can read, great, let them work with that 8-year-old if they have an interest in common) instead of age, that might work much better than any sort of age-based grouping. Of course that can exacerbate the social aspects, but there are in fact ways of making this work well. Having the older student partially teach the younger one, for example, is a good way for the older student to significantly improver their own understanding of the material.
Maybe they should learn a diverse group of children. Girls and boys are ‘different’ but we don’t segregate them anymore...why should we segregate them on the basis of one year age gap?
Except all the places we do (single-gender schools do exist, and even in non-single-gender schools physical education and health are often taught separately).
Also, maybe we should do more of this; there is some evidence in the literature that at some ages educational outcomes would be better with gender segregation if we did things right in terms of keeping equality of resources. Which is of course the sticking point.
> why should we segregate them on the basis of one year age gap
In case it wasn't clear from my answer above, I don't think we should. I think we should group kids by interests and current level (i.e. by what they will be trying to learn) a lot more than we do now.
I am saying that there are some studies showing gender segregation at certain ages may improve educational outcomes, largely depending on how it affects teacher behavior. Whether that means we should do it rather depends on whether we can get those teacher behavior effects in other ways and whether gender segregation would cause other issues (e.g. unequal resource allocation).
For some specifics that are pretty easy to find, see http://econweb.umd.edu/~turner/Lee_Turner_Gender.pdf for recent evidence that boys do may do better in all-boy schools, at least in some cultural contexts. There were a bunch of studies in the '90s that claimed girls do better with no boys in the class, due to teachers actually noticing them, but that effect seems to have more or less disappeared over the last 20-25 years.
In general, as in all things to do with kids and education the answer is almost certainly "it depends". Some children do better in a gender-segregated environment. Some do better in a gender-integrated one. Some don't particularly care. Hence all the caveats above about "some" and "may" and so forth. The hard part is figuring out when gender-segregated education is appropriate and when it's harmful, on a student-by-student basis. Unfortunately, public education is too cookie-cutter for such details.
5 year age difference may be a bit much but beyond that, montessori is doing exactly that. they group children in ranges of 3 years: 0-3,3-6,6-9,9-12,12-15,15-18
we can probably argue about the exact age ranges (maybe follow the schools age-ranges: preschool/kindergarten, primary school, middle school, high school all sound like good age groupings).
but the concept is good. younger kids can learn from older ones. older kids internalize the material by helping younger ones.
and it turns out that this model is already proven too: it's applied in montessori, and it's practiced in scouting as well.
America doesn't need more acceleration, but nor does it need more deceleration, which is what this policy does. What they got rid of was: some kids are ready for Algebra in 8th grade, and they can take it. New policy is: everyone takes the same class, regardless of skill or preparation.
Everyone gets the same 'x' regardless of individual circumstances. Easy policy to prescribe if one holds certain political beliefs.
Unintended consequences: Wealthy and middle class kids will still have access to the same quality of education as they did before, those that rely solely on public education will have access to lower quality.
Inevitably, students from CA will be less prepared for college entrance exams. CA will have to institute a 'statewide college entrance examination' and, maybe they'll pass a law prohibiting using any other entrance exam in an admission process (because they're racist or something or other).
100% agree - the real intent here is to reduce the achievement gap, and the real effect is to increase it. High achieving wealthy parents make sure their kids get math outside of school. The kids who lose are the smart, poorer kids who would have thrived and advanced given the challenge.
Plenty of extremely poor Chinese immigrants scrimp and save every penny to pay for top notch tutoring for their children. This isn't really about wealth, it's about culture and ability.
I half agree. Chinese immigrants strongly value education, and on average, have more inherent ability. So many overcome the disadvantage of being poor. That doesn't mean that the average poor kid isn't at a disadvantage, and that the smart poor kid is hurt by removing advanced math courses. That smart poor kid may not be getting the support at home like a Chinese immigrant, so support at school is all they are going to get.
Why not send them to private school then? Or is top-notch tutoring only available outside of a traditional school? In which case, why have a traditional school model at all?
It is not only that, if you parent is a successful engineer, businessman or whatever. Not only they will hire tutors if needed they can often teach kids themselves.
I learned more math from my dad who has PhD in physics then from all of my teachers and tutors combined.
One of the reason home schooling works so well. It is much easier to teach few kids you care about then whole classroom of the ones you don’t.
From the school district’s page justifying the new course sequence:
“Historically, rigor meant doing higher grade-level material at earlier grades, and equity meant providing all students equal access. The CCSS-M require a shift to seeing rigor as depth of understanding and the ability to communicate this understanding, and to seeing equity as providing all students equal success.”
If the goal is equal success for everyone, you have to hold back the high-achievers so the rest of the group can keep up.
The problem is hinted at here: "equity meant providing all students equal access"
What they are saying is that they refused to keep unqualified students out of the advanced classes. Keeping those students out would have exposed the schools to all sorts of accusations of discrimination, so they didn't do it. Parents insist that low-performing students be in the advanced classes, and the school doesn't say "no", so the class becomes a mess of failure.
Not surprisingly, "You won't be able to do this [computationally hard thing]" is a prediction that doesn't age well. :)
The most recent benchmarks I've seen show HEVC encoding to be 10x-15x slower than H.264, and 4K is 4x the pixel count of 1080p. Naively, this would make UHD 40x-60x more computationally intensive than 1080p, but it is probably not quite that bad. (4x pixel count should be a bit less than 4x computation.)
Latency isn't really a factor here, though. UHD doesn't have higher latency than HD; both just need to fill a few seconds of buffer in the player, and as long as they can keep up with a stream in real-time, this will only take a few seconds to do.
> What do UHD recorders do to record UHD material -- just massive iframe only codecs?
Yes, but the use of iframe-only is orthogonal to the delivery codec or format. (Most pro video is shot and edited in an intra-only codec.)
At these delays you can use sharded coding with clustering. I.e., at a GOP of say 10 seconds, this gives you 8 shards. And with that many shards, I think a nice blade center full of high-clock current-gen CPUs can do it. I.e., 8 systems with as many max speed cores each as the encoder can handle.
x264 for example caps out at about 20~60 cores for 1080p, but I didn't save the exact numbers.
10 seconds is a good balance between seekability and gains due to B-frames, for normal movie content. If the movement is particularly absent/easy to compress, e.g. with screen recordings of Minecraft (as long as there are not too many particles around), one benefits from a much larger GOP, towards the minute and even over a minute. Though you pay for this with most of the seeking you could do before. Choose wisely depending on what it's for.
Well the 300$ Sony action cam in front of me can shoot 2160p30 @100Mbit/s in good quality, considering the form factor. A nicely clocked Nvidia GPU can also use it's H.265 encoder to record at double that, both bandwidth and framerate. These are I/P frame coders, at least you'd not want to use B-frames at these speeds as more effort in your P frames yields more ROI. Intel should, as part of QuickSync, also offer such acceleration, but I am not sure how much that encoder can handle. I am confident it can do something in UHD with P-frames, but I have no idea how good it feats compared to the Nvidia encoder. If the UHD coding is within it's design range, it should yield higher quality though, albeit at barely enough frame rate.
If you try, you can push over 100 fps through the Nvidia one, but quality will suffer even more.
In general, we put a lot of work into performance, developer experience, product flexibility, and support.