HNHacker News
TopNewBestAskShowJobs

mjfisher

620 karma · joined July 27, 2010

submissionscomments
mjfisher··on GraphQL Is a Trap?
This might be just over-familiarity on my part, but does SQL really have a strong learning curve, or is it just not used often enough directly these days that people can get by without knowing it?

Standard SQL is a really simple grammar and a very small keyword set - there's basically selecting, updating, deleting, filtering with where, aggregate queries, grouping and joins, and that's like 95% of it. Sub-queries maybe too.

mjfisher··on It took me 10 years to understand entropy
> The only thing I don't get is how physicists get around the discrete and finite restriction.

Actually, they don't! When you start doing the math about states in a quantum sense (i.e. statistical mechanics), the basic premise is that the available range of states _is_ discrete. Particles are quantized - so they can only possess certain allowable discrete energy levels. The broader laws of thermodynamics fall out of that and appear to be continuous as you scale up to the macro world across a huge number of microstates.

mjfisher··on UK.gov to make adults give credit card details for access to Facebook or TikTok
I believe they're talking about doing that, but contracting out to approved 3rd parties instead of the government for age verification. I still don't trust them to do it securely, and I worry deeply about the chilling effect it will have on the internet as a whole.
mjfisher··on Cavalry – 2D Animation Software for Mac and Windows
While we're on shoutouts and recommendations, I've recently started playing with Wick Editor[0], which is a very good, simple, open source flash-style animation authoring tool. SVG import, mp4 and HTML5 export.

I'm intending to use it soon for some animations to communicate some concepts better over video. The only real limitation I've come across so far is a lack of shape tweening, which would be a very nice to have.

[0] https://www.wickeditor.com/

mjfisher··on Microsoft Irish subsidiary paid zero corporate tax on £220bn profit last year
I don't think that's really such a huge loophole. Pension aside, it's a 2% difference in total tax burden that's probably going to be swallowed up by other company related admin expenses anyway.

I'm self employed and get the bulk of my income through dividends. As with the example above, I often end up paying roughly the same amount of tax as someone with the equivalent salary would.

The big advantage for me is that I can have a very good year and a very bad year and pay an appropriate (smoothed out) level of income tax across both - which seems fair to me.

mjfisher··on Mars Helicopter Lands Safely After Serious In-Flight Anomaly
Silly spacecraft engineers. If only they had the foresight to run "npm install -g ingenuity-camera-api"
mjfisher··on Gromit-MPX Lets You Draw Anywhere on Linux Desktop Screen
I've actually done this a few times before - it works really well until you write on a screen with a particular kind of high gloss anti-reflective coating, which it damages, and leaves you with visible lines even after you wipe the ink away. Not recommended!
mjfisher··on EU bans Belarusian airlines from European skies
That's true - it's also worth explicitly noting that when the EU does make law directly, that is still done in a fully democratic way.

The fact that that EU laws can override national laws does not mean that the EU Commission has unilaterally imposed law without consent. Law changes must first must be approved by the Council - comprising heads of state - and then by the EU parliament, consisting of directly elected representatives from all member states.

I know you weren't asserting otherwise - I just wanted to provide additional important context for those who wouldn't otherwise be aware.

mjfisher··on Freshly made plutonium from outer space found on ocean floor
Or it turns out Bertrand Russell's teapot was made of Plutonium and de-orbited.
mjfisher··on Limiting Work in Progress (2020)
Perfect, thank you - just ordered it.
mjfisher··on Limiting Work in Progress (2020)
If anyone wants a much more in-depth look at the concepts behind this, I can't recommend "The Goal" by Eliyahu M. Goldratt enough. It's a bit like The Phoenix Project, in that it's a novel about improving flow in delivery processes, but it's much, much better, and goes into way more practical depth. You'll learn about not just WIP, but floating bottlenecks, buffers, batch sizes, etc. that turn out to be universally useful concepts. The audiobook is especially good - easily the most practically applicable book in the field I've come across. Go read it/listen to it.
mjfisher··on Learning Ruby: Things I Like, Things I Miss from Python
Not a huge amount in concrete terms of capability, but the little differences add up in how the languages are used in practice. Python has functions you can refer to and pass around directly (as opposed to Ruby's blocks and procs), and that makes functions that are also generators just a little bit more friction free and more likely to be used. Python also has explicit support for differentiating between generators and regular functions in other parts of it's syntax (e.g. list comprehensions) that tend to make people reach for generators more quickly in Python than other languages.
mjfisher··on Is the Schrödinger Equation True?
You might be thinking of Quantum Field Theory, which models particles as excitations in fields specific to those particles. It's completely compatible with the Schrödinger equation, but operates on a higher level of abstraction to allow easier understanding of systems of multiple particles.
mjfisher··on AWS Lambda Terraform Cookbook with working examples
One of the striking things about serverless development that is less obvious from the outset is how it blurs the lines between application and infrastructure.

Deployment of a service is rarely in practice just deployment of new code to an already provisioned lambda - because that lambda can do nothing in isolation. Instead, it tends to be the lambda alongside an SQS queue and a trigger, and an S3 bucket; or an API Gateway that links an authorizer to the Lambda's code. Because of that, evolution and development of the application tends to require evolution and development of those surrounding infrastructure pieces in tandem.

As a result, managing the infrastructure of your serverless service is often most naturally done alongside the application code itself - indeed, the distinction becomes somewhat meaningless. That also means the engineers developing the service require the ability to own and operate the infrastructure as well. That may or may not be well served by Terraform. It's a tool I absolutely love for mutable, stateful infrastructure, but something like the Serverless Framework or AWS SAM can be a much lower-friction and more natural fit for serverless work.

mjfisher··on Write code. Not too much. Mostly functions.
The primary motivating reason to have abstractions in the first place is to prevent context switching - i.e. you shouldn't have to think about networking code while you're writing business logic.
mjfisher··on Ask HN: We have the expertise but no clients. How to reach them?
Even if people don't have particularly accessible LinkedIn profiles, if you know their names it's often possible to guess or construct an email address for them based on the format of other publicly available emails for the same institution.

There are also tools like https://hunter.io/ and https://rocketreach.co/ which attempt to automate the process for you. I've had very mixed results with those kinds of tools, but they're worth knowing about.

mjfisher··on Do we really travel through time with the speed of light?
That's exactly it - it's not wildly incorrect at all.
mjfisher··on An interview code submission that wasn’t even submitted changed our process
I can go one better, and make any program at all compile by removing code with the annoying wiggly red lines under it.

The tools and options we have don't exist in a vacuum; you still need good judgement and practice to make the most of them.

mjfisher··on An interview code submission that wasn’t even submitted changed our process
I commit frequently and rebase to make it nice afterwards - which seems to be a common pattern. For commits that I know at the time need revisiting (partially complete feature/tests don't work/solution works but needs heavy refactoring) I'll prefix the commit message with "WIP" so I have a hint when I come back to it. When the problem is finished, I go back and neaten up the history into a smaller number of logical commits that would make as much sense as possible to someone reading the history.

Tests (in my opinion) need to run continuously during development. I usually have tests run on file change - either in a terminal window or IDE. They also run on each commit when each branch is pushed.

If you have no option other than to run a test suite that takes hours, that needs fixing. You don't need to run all your tests during development, but pragmatically you need to run enough of them to give you some confidence that you haven't broken things. Some options you might have:

* Separate fast and slow tests (e.g. unit tests and integration tests). Run the fast tests frequently as you develop, and use the slow tests to sanity check what you're doing from time to time

* If there are no fast tests (e.g. everything is almost an end to end or full integration test), write some faster tests just for the feature you're working on and run those. It doesn't need to be perfect; just lay down a start that you and your team can improve on

* If all your tests are fast, but there are just a lot of them and they're so abstracted from the layer of code you're working at you can't tell which are relevant, you can add tracing statements to define a subset of tests that hit the codepaths you're changing. Run the small subset regularly.

* Spilt up your codebase - this doesn't have to be into trendy microservices if your problem domain doesn't support the use case. But you'll almost certainly be able to pull out modules or libraries into separate repos that have their own tests. Version them semantically.

As a discipline, we work across a huge range of problems and I'm wary of making assumptions about yours; but slow tests have a massive impact on how quickly you can deliver in my experience.

mjfisher··on Ask HN: Why isn’t visual programming a bigger thing?
> Text-based programming is limited in dimensionality. This can become very constraining

In contrast, I've often thought that visual programming tools are much more limited dimensionally, and that's why they can become difficult to manage beyond a low level of complexity: you only have two dimensions to work with.

With the visual programming tools, the connections between components need very careful management to prevent them becoming a tangled and overlapping web. In a 2D tool (e.g. LabVIEW), you could make a lot of layouts simpler by introducing a third dimension and lifting some components higher or lower to detangle connections - but then you'd face similar hard restrictions in 3D.

Text based programs suffer from no such restrictions; the conceptual space your abstractions can make use of is effectively unlimited, and you can manage connections and information flow between pieces of code to maximize readability and simplicity, rather than artificially minimizing the number of dimensions.

mjfisher··on Driving engineers to an arbitrary date is a value destroying mistake
> Explaining this to a manager who's never even read even one of his copies of The Mythical Man-Month can be tough.

Perhaps he should read both simultaneously and finish in half the time...

mjfisher··on What I wish I knew about React
True, but it's important to remember hooks were only released about a year ago - many large projects in existence before then will either not have migrated or be in a state of partial migration.

I'm glad they're putting the effort into making the API simpler and easier to use to address some of those concerns, but React itself has been around for seven years now - the vast majority of existing developments will have been started using versions of React created before hooks were introduced.

mjfisher··on What I wish I knew about React
When people say "web development used to be simpler", they're probably not referring to the front-end tooling (which has certainly improved) - but when most of the functionality of web apps were built and rendered on the back end, with new state being displayed after full page refreshes or simple AJAX calls.
mjfisher··on What I wish I knew about React
Ending up with HoC on top of HoC is also an extremely salient point. While it's not strictly required, the structure of larger React projects highly encourages you to create or use non-visual higher order components to manage connections to state stores, APIs etc - or make use of the Context API to achieve the same result.

In both cases, you end up with non-UI wrapper components to provide the plumbing in a tree of components that would ideally be entirely presentational. That leads in turn to trying to separate presentational vs. container components, trying to avoid nesting any kind of container component within a pure presentational component to keep it easily testable, creating more HoCs in an attempt to separate concerns, etc.

I certainly wouldn't describe React as garbage like the OP - but I think whether it's the best choice is heavily dependent on the kind of app you're building. The design choices it has made comes with significant challenges and trade-offs and we should not pretend otherwise.

mjfisher··on Agile's early evangelists wouldn't mind watching Agile die
This seems like as good an excuse as any to share a link to "Modern Agile" again:

http://modernagile.org/

I think it's probably a good reinterpretation of the original manifesto that is a bit more fundamental and a bit more direct. I like the concepts a lot.

mjfisher··on Why is Maxwell's theory so hard to understand? (2007) [pdf]
You prompted me to read the actual essay. It was considerably more interesting than the comments here would have suggested; which is not a surprise, given it was penned by Freeman Dyson.

His insight that Maxwell's contemporaries lacked even the language to fully describe what a transformative idea he had is acutely interesting. Once again language both shapes and traps thinking.

He makes the same point with quantum mechanics as well - that we're constrained by not having the proper language to describe the most fundamental behaviour. I think it's fair to say that point still stands today.

mjfisher··on Problem of Time
Just to add to what saagarjha said; there's large parts of the article that just don't seem to make sense on their own. I only have a Master's degree in Physics, and haven't specialized in general relativity - nevertheless I usually have enough background to at least follow along with Wikipedia.

In particular, the "Overturning of absolute time in general relativity" section is very hard to follow, and I suspect would only make passing sense to a reader already very familiar with the subject matter. It has a very stream-of-consciousness feel to it, it doesn't introduce concepts as it moves along, and it makes unexplained logical leaps from one idea to another.

I'm glad someone took the time to start the article and it sounds like an interesting subject; but it could certainly see improvement from someone with the right knowledge.

mjfisher··on What the Heck Is Pyproject.toml?
I think you can certainly use Node as an example of much better developer experience. Even the difference between npm and yarn is fairly minimal, and there is a solid community 'default' in Yarn.

I think part of the reason we ended up with so many trivial libraries and dependencies in the Node ecosystem was precisely because it is so easy to create and publish packages.

I love Python as a language, but working with the packaging even just to install dependencies is like pulling teeth every time you start a new project - nevermind publishing packages yourself.

mjfisher··on 1.1.1.1 for Families
It can be harmful when you're not yet developed enough to evaluate that information (and the motives of the people peddling it) properly. There are plenty of online communities that e.g. normalize and encourage eating disorders or suicide, for example. It's our duty to educate our children well enough that they'll eventually be able to make smart decisions themselves - but that takes time and guidance.
mjfisher··on On Scaling Mental Models
I think that was more or less still the point - more expressive power (including custom key bindings) allows you to build more complicated abstractions with less obvious behaviours. It's more difficult to maintain a mental model of those less obvious behaviours.
← PreviousPage 4 of 7Next →