myhrvold

A less nefarious reason behind companies increasing profit margins, is that with more uncertainty -- i.e. potential supply chain disruptions in the future -- companies feel the need to make more from what they are able to get to the end consumer/customer.

Generally speaking, there's a lot more that can go wrong than go right in terms of getting stuff to people especially from overseas. (True anywhere in the world.) And the specter of uncertainty basically elevates the default cost of doing business for everything.

Unsurprising to see TV writing overtake film writing in the last decade plus; TV writings lends itself much more to streaming shows (which are considered TV even if they never appear on cable / "normal" television).

Interesting that adjusted for inflation however, writing pays roughly what it does in the mid 1990s per the article.

That's a little surprising.

It's also probably a more competitive industry however, and the median pay (which is what's measured) doesn't account for the best writers probably doing better than ever. And then maybe more part time people trying to break into the field but not getting much.

If there's a pretty large middle ground buoyed by union pay scales and contracts, then both ends (high/low) will not really be accounted for in the statistics.

Shopify still a valuable business! It's a success story in a volatile space. So if they want to maximize their growth during good times, it's almost inevitable they'll have to walk back some of the earlier steps they made during fallow times.

So I would put Shopify in a different league altogether than RIM or Nortel... these latest moves in fact may help them avoid the fate of the others.

Unfortunate, but these layoffs are not surprising given economic conditions.

It's a sizable number of affected roles, so hopefully this is a one-and-done for now. (A drip, drip, drip over a year and a half is the worst IMO with respect to morale and uncertainty for the remaining folks. And then you end up losing some of the people you want to keep because they view their employment environment as unstable...)

Smart! This was a bet that I think made sense in theory for Shopify, but in practice became too hard to manage and execute given their M.O.

Would imagine Shopify is taking a loss in offloading, but will prevent future hemorrhaging of cash from a business decision that clearly did not pay off like anticipated.

Unlike some people, I don't think this experiment was necessarily a bad one for Shopify. I would have to know inside info as to whether it made sense at the time to do, but barring that I would give the benefit of the doubt to Shopify to try, if the reward was high enough if they pulled it off...

Nope that was nearly a month after -- toward the end of February as I recall. #DeleteUber started when Uber turned off surge to an airport from a taxi strike. It was headline news already by the end of January: https://www.nytimes.com/2017/01/31/business/delete-uber.html And people somehow thought that getting rid of surge temporarily was a bad thing, or using that message to take advantage of the situation. (I don't know exactly what the rationale was, because it still doesn't make sense to me after all these years...)

Nah the taxi industry was a lot worse than Uber. Most of the arguments over whether Uber drivers are employers to me are baffling; most taxi drivers took set, long hourly shifts from the medallion owners -- a much greater hallmark of an employee/employer relationship -- and were mafia backed and run quite crookedly. Many taxi drivers to this day have much worse working conditions and pay rates.

On the rider side, many people of color, especially Black, were not able to get taxi rides because they drove right by them. Look this up if you don't believe me -- there are innumerable examples of documentations of this, and this basically was altogether eliminated with Uber rides.

Uber was a game changer and a great equalizer for a great many people in society who didn't own their own cars (i.e. visitors, but also people who lived in major cities.)

In fact many taxi drivers became Uber drivers and have made a much better living this way.

Of course, there are also many more Uber drivers than at the start... and so it is kind of a race to the bottom with respect to an equilibrium achieved where it's now just an OK job instead of the $30+/hr like in the early days for uberX drivers...

A lot was also in a gray area. In fact France ending up (I think very unfairly, but ofc I worked at Uber) passing laws specifically against Uber to make it retroactively illegal. Which basically was a tacit admission that is was OK by the rules initially. Obviously the idea of being able to "hail" a ride -- specific to you -- in near real-time was not really considered around the world when a lot of taxi rules and regulations were done.

So you ended up having insanity, like the necessity for even a black car service to "return to their base" after a ride, for time periods up to 1.5 hrs. Basically trying to nix the notion of efficiency in ridesharing.

We all know that nowadays most of those rules were battled back.

But the taxi industry and regulators had some very crooked ways which were really to the detriment of broader society trying to get places without having their own car.

Yes the beginning was in late January 2017, involving a taxi strike at a NYC area airport and Uber's messaging around continuing to operate to serve the airport and surge pricing. That's how the delete Uber hashtag arose and it just continued with wave after wave of negative stories for the next few months.

The whole way things unfolded subsequently (due to several other pile-ons in following weeks, and eventually TK's ouster in mid 2017) still seems surreal to this day.

(I worked at Uber for ~5 yrs and remember this time period well.)

Yeah I think that's in part because so much of the cultural zeitgeist around Uber was on the policy implications, and the effect on consumer behavior etcetera.

There were a ton of technical challenges in scaling.

Uber was also one of the earliest corporate deployers of Node back in 2010 and ended up having one of the largest in the world arguably (have no idea to what extent they've replaced in the last few years).

Uber's open source projects have led to spinout companies that have raised a lot of money successfully like Chronosphere and Tecton.ai .

Halloween and then New Year's were huge traffic challenges and the Uber app (both driver and rider) really pushed the bounds of what was possible in the 2010s with respect to a two-side real-time marketplace for ridesharing and food delivery.

(I worked at Uber 2014-2019 in engineering.)

Inaccurate overall IMO; a caricature of what actually went on. But accurate in terms of capturing the sentiment that a lot of Uber skeptics had during the height of Uber mania, say 2015-2019. So the series is more a reflection of that than a faithful accounting of Uber's rise.

uChat actually saved Uber a lot of money by the way. Slack is way more expensive and it would not meet the requirements of Uber at the time, i.e. anyone being able to chat with anyone (including the famous global Pool Party room with every single employee in it). Also Slack had an issue with Uber's scale, having messaging alerts for services that would not reliably show up that engineering would use to detect outages.

uChat was based on Mattermost as I recall which was open source, and saved Uber several million dollars plus over its lifetime course.

I was not directly involved in its development but I (started and) ran the engineering blog, and also Uber's open source program for a time where this partnership occurred.

It is funny how some of these examples are floating around about build vs buy, but uChat actually was a good deal while it lasted.

Ultimately Uber decided they didn't need everyone talking to each other all at once and so bit the bullet and paid more for Slack.

Maybe. The buyer certainly has more info for how market valuations are set if they're an investing professional, like is often the case for these tender offers.

However, on the other side... you have long standing employees of the company! So they are by definition insiders. And you'd have to think that some subset of employees actually would have info about the company that buyers in the tender offer process wouldn't have, particularly execs or in finance or maybe some key component of technology that the company's ultimate success hinges on. (Else, you wouldn't see public company restrictions on employees trading shares.)

I believe the general thinking around tender offers is they're often done on some level of discount on what an "open market" offer could be, at that point in time. In part, that's because private companies like to control their cap table. And so you're right that this dramatically reduces the competition in the bidding process as basically it's large funds/firms. Of which there are way fewer than, say, retail investors and smaller funds who would be willing to buy fractions of a few dozen employees' stakes.

Nevertheless, many people taking tender offers are not necessarily doing so out of desperation for liquidity per se. It's an opportunistic way to get some value out of your having worked to build a company, besides salary.

Yeah Uber had this, twice actually as a private company; once with Softbank in January 2018. And then again with Coatue / a few other investors in July 2018.

There's a bit of game theory at stake, as to what you tender. You want to cash out a certain amount to diversify... but if too many people say yes, then your stake is whittled down to due to oversubscribed interest on the selling side.

You get one shot to put in for your % number, and it will either be that, or proportionately lower depending on every other eligible employee's interest in selling.

I recall the max was throttled at 50% vested shares, or something like that anyway for Uber. And you had to have a minimum number of shares vested to participate (low tens of thousands overall as I recall?)

Anyhow, I put in for the maximum both times which worked out because that % got whittled down since the tender offer was filled on both occasions (as I recall).

(P.S. Not that it matters for the story, but $40/share in July 2018 and just under $33/share in January 2018 ended up working out great compared to what happened subsequently.)

Stripe made the right move here. Going public is a huge move. You don't want to make a big decision like this because early employees couldn't have cashed out.

There are lots of other considerations that come along with an IPO and Stripe is addressing the core problem (early employee liquidity) with a targeted solution.

(And yes -- they would've been better off doing two years ago. But timing is hard, and that's not really relevant to their decision now, which is the best at the current time given the past.)

Yeah I think the question is "great for who". I think it could be great for the founders (who can achieve liquidity in a variety of ways at any point really, that regular employees don't have access to. I'm not critical of this -- makes sense to me -- just pointing that out.)

It wasn't so good for employees who would've wanted to sell and diversify when the valuation was, conservatively, about 2x what it is now. And maybe even 2.5x plus if you extrapolate what Stripe could've been worth at the height of the market.

For the long term company, it still could be a good, because if they can raise capital privately like they were just able to do, then they get the upside of access to money without the downside of share price fluctuations, scrutiny, more bureaucracy etcetera.

Yes; the idea is that the startup wants control for who will be on the cap table, so they don't hassle with written requests for info etcetera. Minority shareholders (especially for California businesses) have a fair number of information rights, and so companies tread cautiously in terms of who they allow to invest. This is in part why special purposes vehicles exist so that there's a point of contact for investment. (And then the company doesn't have to vet people that invest in that syndication as much...)

That's true but the reality for many will be selling at least the amount of shares to cover the cost of executing their options. And because the next window to sell could be undetermined (possibly many years in the future), many employees will find it prudent to get liquidity that would cover their cash needs for the next fair-number-of-years. (Based on their understanding of what their stake is worth now. Of course they're going to have that number in mind even if it wasn't able to be realized until now...)

A target audience is a good broader point about MOOCs and online education. (I took edX courses during the "glory days" of 2012 and 2013 -- and also tested out Udemy, Coursera, and Udacity at the time.)

The one-size-fits-most nature of online education goes against the "customize your education at scale to learn" which was an earlier anticipated advantage about MOOCs. Specifically, adaptive learning and being able to accommodate a variety of learning behaviors and styles. "Learn at your own pace, in your own way, on your own time but still within bounds to the rest of the class" kind of thing.

I remember when Stanford launched online CS courses in the mid 2010s, that it was thought they'd have the best of both worlds and their in-person, offline course offerings wouldn't be affected. (Diluted down to the lowest common denominator of student, which now included online learners who weren't Stanford students per se.) Well, over time, turns out double duty-ing course material for online and the "regular" classes crept into all education for instructors. Which meant the courses with online equivalents became easier across the board. Thus, the target audience for everything shifted.

Again, with acknowledged intentionality, I don't really have an issue with this (which you could crassly summarize as "dumbing down" the course offerings for convenience's sake) -- except that from my vantage point it was an unforeseen consequence of part of the online and MOOC push.

Obviously, I benefited from online courses in my mid 20s and so I look at their rise with nostalgia and through a rosier lens than many. However, I also can't help but think that they ended up being not quite what was promised at the outset, which was better targeting in addition to expanded educational access around the world. Especially for students who thought they'd signed up for the more challenging materials and didn't want to be part of a grand new experiment.

Interesting that MIT OpenCourseWare will outlast edX, which for a few years truly did look like it was the future of university level education and beyond.

The way I read it... the Wayback Machine did not allow archives to be taken down arbitrarily -- but a subsequent targeted robots.txt exclusion of the Wayback Machine could render prior archives of that website moot? (Because the Wayback Machine starts from scratch each time?)

Also: Tender Loving Care or Taxi & Limousine Commission :0p

Norway's currency, the krone: https://en.wikipedia.org/wiki/Norwegian_krone (They are not on the Euro.)

Blue Apron comes with its own meal kit and instructions on how to make. It's almost like one of those DIY science experiment boxes, but for dinner. This makes it more appealing than the package of what a grocery store could offer. (Or selecting the individual ingredients one by one.)

For Blue Apron, you cook + assemble the meals yourself which does not sound too different, but could be a huge advantage for cost (and quality standpoint vs you expect it prepared well and warm).

Thanks! Figured there was too much traffic when the main link was unavailable a few min earlier. Was wondering whether it was a direct view of a notebook like via nbviewer (in which case, would be ironic.)

I've never heard of him espousing detrimental views FWIW. He had a public spat with Google after he left though: https://www.quora.com/What-does-it-feel-like-to-be-fired-fro... There used to be a lot more blog articles about it (on that same blog I believe), but now that it's been a number of years (maybe even 10 by now?) I think they've slowly started disappearing.

Edit: here's the explanation on why you can't find a lot of his earlier material online: https://www.reddit.com/r/programming/comments/44bmmq/why_agi...

Well, yes, but the acquisition in Sep 2016 (referenced in another reply to this comment) was just over the IPO price: https://techcrunch.com/2016/09/08/google-will-acquire-apigee...

Facebook had a similar valuation decline, then a bounce back.

So those who had faith in the company for another year saw modest gains compared to 1.5 yrs ago at IPO.

Yes -- unlimited vacation is a perk of flexibility rather than number of days off.

Since you're responsible for getting your work done, you can take any manner of time off provided that the results hold up. This past year, I prolly tallied anywhere from 4-6 weeks off, technically speaking. Unplugged though, a handful of days at most. (Otherwise, mostly on vacay but maybe a few hrs worth of work a day sprinkled in morning / evening.)

Because I own my area of work + results though, during decent chunks of this time I still answer emails when I can, try to unblock people etc. So it means that I get to travel more and see other places, or work in other offices (would say 'remote' here but some are in places like NYC & Seattle so hardly remote in that sense of the word!)

Then, when I am around town in San Francisco, I make a point to check in with people in person, often through the evenings. And work part of the time on Sat/Sun to unblock people if I need to review or provide someone else the context to do so.

For my role I have made this work and love it, but this unlimited option is not for everyone. And there is a cap to how much you can realistically be away and still get things done.

Seems like in this case, how the company originally presents what you're getting matters. So, as you point out here, qualifying the # of days off w/ days that are set [that count as part of that allotment] would set expectations accordingly. Otherwise, still a good deal, but employees would feel misled if they want to use it other parts of the year.

Link appears broken as of now, FYI... (1:00 AM EST)