blacktriangle

Please don't. He's doing great work with Brave and I'm enjoying watching Mozilla commit suicide.

How much of a facade would you have had to build to be useful to the point where you run into the same problem? It seems like it would be very very easy for implementation specific behavior to leak this way.

At least with IndexedDB there's a large enough impedence between SQLite and IndexedDB that there's far less chance of details getting through.

They were opposed to having a web standard, which should be a generic standard, be implementation defined. Mozilla would have had to re-implement SQLite for WebSQL based on the principle that there should not be a standard with only a single implementation.

There's nothing wrong with implementing the actual standard IndexedDB on top of SQLite, since it's entirely possible to reimplement IndexedDB on top of a different underlying database.

WebSQL would have been awesome right up until the moment that some new platform takes off where SQLite won't meet the requirements for but we want to reach with the web platform, at which point we'd be screwed. Or even worse, WebSQL would guarentee that the reach of the web platform has been fundamentally limited by what a single codebase can reach.

Yep, it's not fear of climate change, its fear of giving up on Tinder.

Clearly crazy person throws out unsubstantiated complaints along with a bunch of complaining about non issues, and you think that's justification to put a company under scrutity? Ok commissar.

The problem with "startup" is that there are two separate issues mixed up in the same word. PG defined it as hypergrowth which is certainly one class of startup, but there's another one as well. The other aspect of startups is that they are newish products doing something new in new markets. As such, issues like product-market fit are key. The problem is when you start to say things like you're not a startup, you're a traditional business is that most traditional businesses don't have the product market fit issue, and thus most traditional businesses are far far less risky. For example, I don't need to convince a bank that my coffee shop will achieve product-market fit, we know people love coffee and have lots of tools for evaluation the financials of a potential coffee shop, which is why I can get a bank loan to go start up my shop.

Now there's the issue of starting a company that is trying something new and lacks product-market fit, thus is higher risk and we don't have good tools to evaluate those risks, but is also looking to grow conservatively thus isn't able to get the interest of those looking to give out high-risk funding, ie VCs.

Oh noes, he has ammo at his desk. Come to my state lady, pretty sure everybody but one of the people in my last office open carried.

But now you have to look at the expected value as well.

A 10 million dollar company, lets say at a conservative 5x earnings multiple means 2mil annual profits. At 80% margin that's 2.5mil ARR. For a B2B SaaS product you should be able to get at least $1,000 / customer annually, which means you need to find 2500 customers to own 100% of a 10 million dollar company, which in the age of email and Facebook marketing is very much within reach.

And these are conservative numbers. A strategic buyer might very well pay 10x if you're showing nice growth, margins of 90% are not unrealistic in SaaS, and you could possibly raise prices depending on the value you are providing and who you're selling to.

Now ask how many stars have to align to reach a billion dollar valuation, assuming you haven't been screwed over by your investors by the time you reach that point.

FWIW I'm working on a tech startup far away from SF, so it's possible. However we're arguably not a startup since we're not really engineered for hypergrowth as much as slowly and methodically building a self-funded, profitable product company. As such issues like raising funding and hiring quickly are not things we're concerned with, which might be why we're doing fine outside of SF. When we do hire it's 100% remote so our location isn't hurting us in that regard.

We feel like this is the best choice for our company as we retain 100% ownership and can focus on a straightforward business model of selling people a product that solves their problem at a fair price. Arguably the possible upside would be higher if we tried to go the VC route and get huge fast, but then the potential downside goes up as well. This way we're working ourselves into what amount to incredibly high paying jobs while also owning the asset that is the company.

Same goes for Emacs, nobody points you to the Info system, which is absolutely amazing.

Perhaps this is because both vimtutor and the Emacs tutor do refer to their respective help systems, but given the sad state of most help systems beginners are familiar with I think they can be forgiven for not taking the recomendation seriously. Its worth really emphasizing to beginners that there is a follow up to the simple vimtutor/TUTORIAL built right into the editor.

I knew an older, highly successful CXO type who had changed careers from developer into management and gone to a low-tier MBA school in the process who had a hilarious line: "I've worked with a ton of Harvard MBAs, I had to fire every one of them."

Your point about career shift is really where the MBA shines. MBAs should be viewed as a tool for those who are moving out of a direct technical field into management, and MBA students should bring some real world context with them to their classwork.

The problem with the so-called elite MBA schools is they are designed not for people with experience, but recruit those right out of undergrad with 0 business experience. This results in a general graduate population with highly inflated sense of confidence and no actual work to back it up, a very dangerous combination.

Urbanization could be a huge part of it. In my rural area, it's pretty standard that retired men universally convert into avid hunters. I'm learning that no matter the time of year there's always something tasty you're allowed to go out and kill.

The timing is huge. There was a period time from the late 80s to I'd say Win98 when computers became affordable, they were useable by laypeople, but they still weren't easy yet. You still had to learn about some of the underlying ideas to do anything useful. By 2001 user-focused OSes had gotten figured out to the point where they mostly worked and people needed to know less and less about the underlying technology.

The key really was that brief time period where hardware had finally become widely available but software was still catching up with the new mass market.

Or this is the inevitable end of leftist thinking every single time. Sounds good on paper, fairness and all that, but totally ignores reality and ends up requiring large amounts of violence to enforce its idea of equal outcomes for all, where that equal outcome is always poverty and starvation.

The Facebook whistleblower is a clear government plant sent there to justify government take over of Facebook's popular opinion control apparatus. The Republicans want to shut down the faction that want Trump back and the Dems want to shut down the Bernie bros and Yang's new party, so both parties are united in their desire to regain control of public discourse. When's the last time you've ever seen a whistleblower treated so kindly by the feds?

it's a bigger issue for developers. The fact that I can do *nix style development, then run some script that wraps it up in a nice .app bundle and ship that off to users means that its popularity very much matters.

Take a small example. Is Figma replacing photoshop for designers? Is Google Office replacing MSOffice? If these were to fully replace their local counterparts, then they would also obsolete the whole concept of navigating a file system as well.

Sadly one way to look at this might be similar to "the number of people who can drive an automatic transmission and change their own spark plugs is decreasing." Its true, but the underlying tech is changing too, so is it really a problem?

I have a theory about the Japanese and their relationship to the PC, having lived there for a bit in the mid 90s. This comes down to three things:

First, Japan had a mindblowing consumer electronics industry way before we did. There were special purpose devices for everything you could want. They were all user friendly and did not require understanding how the underlying system worked. They didn't need a PC since all their electronic needs were already met and met better.

Second, Japan is basically 99% urban living, as space is at an insane premium, and even when you are in the country the houses are built small due to culture and lack of open space, since if you're not in the city you're probably in the mountains. This means there was no physical place to even put a PC in most Japanese houses the way American houses might have a desk in a corner or even an office.

Third, and probably the most important, phone calls in Japan were always pay by minute, even for local calls. This meant that if you were online with a modem, you were paying the phone bill just sitting there. This meant that if a person even had a modem, the would rarely use the internet. Since America had monthly-billed phones, there was no problem staying up all night on AOL chatrooms, and AOL really was the killer app that pushed PCs into American houses.

IMO this is why the PC revolution that really took off in American homes, and by association a bunch of us who cut our teeth mucking about with bat files so we could play our DOS games and ended up learning useful skills out of necessity never happened in Japan.

We have one in our app, I've been amazed how users have reported so many bugs, feature requests, and tiny issues through it. Chat widget was totally worth it.

I just read that as "fiat currency has failed."

And at this point you may as well be running Datomic. I remember the first time looking at Datomic the whole thing felt like it was coming out of left field. A few years later of data modeling conversations like this and you start to see Datomic is really every fix to the relational model taken to the extreme.

It also explains how we got column databases. If the problem with 6th normal form is the underlying storage, then change the underlying storage to optimize for 6NF.

In a way, the very existence of the normal forms can be taken as a sign that something is seriously wrong with the relational algebra given how easy it is to create a data model that will produce rubbish for certain queries.

If only my pocket WhatsApp device had other ways to contact people, now I'll have to go get a new device.

Seriously as transitions go, leaving WhatsApp and Messanger will be the smoothest most painless transition there ever was.

Sounds like a restatement of the open world / closed world debate.

There's a saying "to make an omelette you have to break a few eggs." It's a justification used by those who think they're bringing about some great positive change to justify the all the damage they caused.

For example, "In order to bring about our communist utopia, we're going to have to murder a few million people."

Maybe trivial was the wrong word, but your second paragraph is exactly why I reject the approach to splitting out all nullable values into their own relation. Yes from a data modeling theoretical perspective it works, but from a pragmatic perspective having to join everything back together again is a worse situation than just thinking through the behavior of null in that situation.

But even then things are awful. What do I do when I want to include the value of the nullable table in my result, but I want all of my records, not just the ones with non-null. So even though you've purged null from your data model you're still bringing them back with an outer join.

Heck I wish he even had any understanding of the use of NULL in real-world databases. He spends books arguing, very compellingly I might add, against the use of NULL in SQL dbs, but when asked to come up with an altenrative to NULL, he's suggestions are sentinel values and splitting out every nullable attribute into its own table. Both of these solutions are trivially rejectable.

Not a part of the SQL standard, but would PostgreSQL enumerated types be what you're thinking of?

https://www.postgresql.org/docs/current/datatype-enum.html

Although I guess the question is are you thinking sum types at the attribute level or sum types at the relation level? At the relation level I'm not familiar with an answer, but I can see where that would be a nice database-level solution for polymorphic associations.

Lets just all go back to calling them mentally ill.

Facebook is no stronger than Myspace, Friendster, Orkut, or any other social network. Everybody looks unassailable at first, but when social networks collapse, it happens overnight and you never see it coming.