HNHacker News
TopNewBestAskShowJobs

pbiggar

7,962 karma · joined January 26, 2009

Cofounder of Tech for Palestine (https://techforpalestine.org)

Previously founder/CTO of Dark (http://darklang.com), and founder/CEO of CircleCI (https://circleci.com). PhD in CS.

https://hachyderm.io/@paulbiggar

https://bsky.app/profile/paulbiggar.com

https://twitter.com/paulbiggar

https://blog.paulbiggar.com

Contact: paul@techforpalestine.org or paul.biggar@gmail.com

submissionscomments
pbiggar··on Ask HN: What are some of the best podcasts for developers?
We haven't recorded one in a while, but I have a podcast called To Be Continuous [1] with Edith Harbaugh from Launchdarkly. It was less about specifics like discussing generics in Go or garbage collection, more about how software gets made (product, startups, coding, devtools, etc). I won't claim it's the best software engineering podcast, but developers have told me they liked it :)

[1] https://www.heavybit.com/library/podcasts/to-be-continuous

pbiggar··on Edge-compatible Serverless Driver for Postgres
I think you're telling me the answer but I amn't sure.

Suppose I use 10TB of storage, and then later I delete all but 100GB of storage. Will I be paying for 10TB or 100GB?

pbiggar··on Edge-compatible Serverless Driver for Postgres
Does Neon scale down storage? I'm paying GCP a lot for storage I don't use, and it's incredibly frustrating - they scaled me up no problem but won't let me scale down.
pbiggar··on Sperm counts worldwide are plummeting faster than we thought
While the trend has been around much longer than Covid, it looks like Covid is going to make it a lot worse! Sperm count is down 53% [1] for people who had covid, and it seems 94% of Americans have had Covid thus far.

[1] https://onlinelibrary.wiley.com/doi/10.1002/jmv.27971

pbiggar··on Tell HN: If you have trouble focusing on projects/meetings, see a psychatrist
Similar thing for me. Went to see a psychiatrist for memory issues, she diagnosed ADHD, and a bit of ritalin helped me focus a lot.
pbiggar··on Dynamic types have the potential to be more than "no static types"
Completely agree. I spent a lot of time thinking about what programming language features are important for understandable code when I was designing Darklang. One of the major negative influences was from Rails, where I had often found it hard just to understand where a particular piece of code came from. Clojure is similar - with "why is there a nil here" being the challenging question. With Python, I made my own mess making things far too polymorphic.

But I had previously had experience with OO languages - Java and C++, and they often had similar challenges, though not quite as pronounced, along with problems of their own.

Eventually, I came to realize statically typed functional languages were the panacea. Sum types offered better composition than OO, and they had a good trade off between power/succintness (coming from higher order functions) and program understanding. In particular, the problems from other languages: what code is this, what type is this, where did this nil come from, were solved.

pbiggar··on Ask HN: Sites like Hacker News but in other languages
Lemmy is a fediverse clone of reddit, and some of their communities are in different languages: https://join-lemmy.org/instances

The "main" site also has many subcommunities in many languages: https://lemmy.ml/communities

(ps I literally found this an hour ago, so I don't know much about it)

pbiggar··on Paredit 25 released, after 8 years
Paredit is so good - completely changes how you edit. It was also the primary influence for making the Darklang editor a structured editor, though I confess it is not yet as good as paredit.
pbiggar··on Stripe laying off around 14% of workforce
- 14 weeks severance

- 2022 bonus and PTO paid out

- accelerated vesting

- 6 months of healthcare.

This is a phenomenal severance package and I hope one that will set the standard for companies doing layoffs. So many companies in the US do two weeks or less, with nothing else (not even healthcare) or even use it to claw back shares.

pbiggar··on Rethinking Virtualization for Backends
This is really cool. In a sense, this is also the approach we took with Darklang. We are running the runtime layer (http servers, DB drivers, OS runtime, etc) so all that needs to be deployed is the core of your app, as opposed to your app plus your language runtime plus your OS etc etc.

I'm curious what shuttle does once your app is deployed? Is a container cloned with the app injected, then connected to a load balancer and left running? In Darklang we just keep the AST in the DB and fetch it each request, but at some point we'll want to do better.

pbiggar··on Replit Mobile App
Lol, thanks for asking! Yeah, I think we're making progress. There's a lot to do, and I feel it is getting done (see progress at https://darklang.com/changelog).

I don't believe we have product market fit yet, the plan goes like this:

- rebuild the technical foundations (DONE)

- fix a lot of the jankiness in the editor <- we are here now

- make sure language is fully baked (coming soon)

- add a package manager/user management

I think the last step is where we'll start to see some product market fit.

pbiggar··on Replit Mobile App
Very very cool. Have been thinking about this for Darklang for a while. One thing that's changed since I started thinking about it is how big a deal AI assist is.

Originally, I thought a structured editor would be essential, but now while it's helpful, I think AI assist can do like 90% the benefit you get.

pbiggar··on Why OCaml?
While you're not wrong about docs, the new package documentation for every package in the ecosystem is pretty fantastic! https://www.ocaml.org/packages
pbiggar··on Why OCaml?
I spent about 20 months porting darklang to F#. It took much longer than I expected. Here's how it went: https://blog.darklang.com/backend-rewrite-complete/.

It probably took longer than it had to: 30% of that time was devops, I wanted it to be bug-for-bug compatible so I spent a lot of time writing tests, and we were using some stuff which wasn't mature yet (new JSON serialization lib, compile to webassembly, binary serializers)

F#/.NET certainly has everything needed, and I find it a much simpler language than OCaml (eg no module games) - really it's what I hoped OCaml would be.

pbiggar··on GitHub Actions Pitfalls
> my chance at a growth industry

I think your view on of crypto might be skewed by this. I think working in coal or surveillance tech or fracking is bad, and that isn't changed by the fact that it is someone's best chance at getting rich from it.

Note that most programmers didn't get rich out of social media or the boom from the last decade: maybe 100,000 became millionaires out of 20 million programmers, most of whom had generally middle class salaries

pbiggar··on Kazakhstan’s president: We’re moving full speed ahead toward reform
I went to dozens of BLM protests in 2020, including the one in SoHo that got the most publicity. The number of looters from among the protestors rounds to zero.
pbiggar··on Blocking Kiwifarms
It took literally half a second to come up with mainstream discussion of this rhetoric. Here's the first Google result: https://www.nbcnews.com/nbc-out/out-politics-and-policy/groo...
pbiggar··on Blocking Kiwifarms
> child predators

What's happened here is that they (and you'll see this has happened across a lot of the extreme right in America) have started to label trans people as "pedo" or "groomers", in an effort to make it seem like trans people are really child predators, which makes many people more tolerant of their attacks and more sympathetic of anti-trans policies such as bathroom bills.

pbiggar··on Ask HN: How do you and your spouse handle big income differences?
If it's helpful, the government* considers it both of your money. If one of you racks up debts that money will go to pay for it. If you divorce, you'll split it. You two are a package deal, so all this pettiness isn't helping anything.

Use some of the money to go to therapy, both individual and couples. If you go soon enough you might save your marriage. Either way, it's probably the best investment in yourselves that's available.

*Depends where you are but certainly true in California

pbiggar··on Mold/macOS is 11 times faster than the Apple's default linker to link Chrome
FYI: https://www.amazon.com/Linkers-Loaders-John-R-Levine/dp/1558...
pbiggar··on I'm as shocked as you, but the Darklang back end rewrite is complete
Yeah, absolutely fair. That said, I think there's a movement away from local because:

a) deployments are getting complex enough that it's hard to get dev environments locally

b) it's cheaper and quicker to pay for cloud development machines anyway

c) things have enough scale that local development doesn't match production anyway

Saw this on twitter recently which supports this a bit:

https://twitter.com/swyx/status/1533910738942562304

pbiggar··on I'm as shocked as you, but the Darklang back end rewrite is complete
Thanks! I understand the wariness - the idea is that Darklang is solving a particular problem, which is building cloud backends. I believe that adding in other constraints (common ones are local development, using existing languages, using existing editors, and using existing hardware/clouds) will make it worse for solving the core problem. Plus, there's plenty of great options that run on your local machine, so just don't use dark for that.
pbiggar··on What did Earth look like X million years ago?
It shows the dinosaurs in a very light grey over the white background. You can find it just below the city search box.
pbiggar··on Ask HN: Have we screwed ourselves as software engineers?
You're not wrong. The problem is the fancy stacks actually solve some stuff, but they bring their own disadvantages with them. No-one is denying that kubernetes is extremely powerful and you might need that in some use cases, but then you're suddenly writing operators for it, instead of business logic for your customers.

I think the problem is that things we're building stuff to solve specific problems, and then expanding each of those tools until they become massive and need other tools to help them. So docker solved a problem but then it created problems that you need kubernetes for, and so on.

One of the reasons I'm working on darklang is that I think the root cause of this complexity is solvable. The solution, in my opinion, is to build tools that cover multiple layers of the stack - that removes join points where you might be tempted to customize.

For example, firebase covers multiple layers, you might otherwise need a DB, a connection pooler, a firewall, an api server, an autoscaler for the api, a load balancer, etc. But instead, the only surface area you have is the firebase API. There's lots of similar tools that cover multiple layers of the stack like this, netlify, glitch, darklang, netlify, and prisma are some examples.

pbiggar··on The great VC pullback of 2022
This is good advice. However bear in mind that the moment a term sheet arrives, the old 409A is invalid. So if the company is fundraising, make sure you get your options allocated immediately (even if that requires the board to have to sign off outside of their regular meeting cycle). Otherwise, once the term sheet arrives, you will get your options at the new strike price and there's not much the company or you can do about it.
pbiggar··on Ask HN: Where can I see many examples of real companies' software architecture?
All of darklang's infra is source available, feel free to read it. This [1] is a good entry point for the infra configuration and setup.

[1] https://github.com/darklang/dark/#production-services

pbiggar··on Advantages of Monorepos (2015)
Monorepos are also great for small monorepos with just a few projects. The darklang monorepo [1] has a devcontainer that installs all the build tools for 4 projects which create 21 different services, using 6 languages, and building everything is one step.

In fact, it makes it so easy to add new stuff that I didn't even realize we had 21 services til I counted. My first guess was 12.

[1] https://github.com/darklang/dark

pbiggar··on Two Vexing Problems in Functional Programming
In addition, the O(logN) only comes into play if you are truly accessing only one item. If you're accessing more you are either reading the whole list (O(N) in total, O(1) per item), or a sublist (in which case it's O(logN) once to get/update the k item sublist and O(k) to operate on it).

If you are accessing just one item, the O(logN) cost is likely dominated by some other cost (eg the user clicking a button, making a web request, etc).

pbiggar··on Two Vexing Problems in Functional Programming
I've been coding in functional languages for 10 years, and they have problems, but these are not them.

The first problem is solved in every functional language I've used (clojure, F#, OCaml) by immutable data structures (aka purely functional data structures, which is also the name of the de facto book on the topic, which is excellent but also you don't need to know any of this to use them. [1]). I'm surprised to see this critique without a mention of this.

The second problem seems to be due to trying to do OO in FP. The thing is you don't have a glyph, you have a whole system which contains glyphs. So you can't do an update on a glyph because it won't update the system.

This is a common frustration when you're used to OO and can be really annoying in FP but it's also part of the design modelling that FP forces you to do: the problem was that he was modeling in the small and not the large.

The solution is not the 1, 2, or 3 from the article (1 and 2 are bad, 3 is not as bad but not FP), but is straightforward. When you want to update a glyph, you use a function like `updateGlyph(pathToGlyph, functionToExecuteOnGlyph)`. That's at the top-level, and it allows you compose your lower level `updateGlyph` functions.

[1] https://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.64...

pbiggar··on Updates to Google Cloud’s infrastructure capabilities and pricing
My $10,000/month bill just went up by $4.41 from "List Price $ increase in monthly bill due to multi-region egress"
← PreviousPage 10 of 34Next →