StanAngeloff

EDIT: resolved.

Okay, the other day it was Claude across providers, today it's v0.dev. AI has become so integrated in my personal flow that an outage literally causes me great anxiety. I hadn't realised how dependant on our new AI overlords I've become.

AWS Bedrock or Vertex AI. I guess https://openrouter.ai/anthropic/claude-3.7-sonnet/providers might be worth exploring, too, as it may provide an easier switchover in case of a failure at one of the providers (I haven't used it myself).

EDIT 2: stable again after lots of intermittent errors.

EDIT: re-opened https://status.anthropic.com/incidents/3j3zbrx1cfbt

Still occurring in Europe despite being marked as resolved. There was a brief period where things stabilised, we are back to intermittent `{"type":"error","error":{"type":"overloaded_error","message":"Overloaded"}}` responses.

Shopify's relentless push towards GraphQL is starting to feel like a never-ending treadmill for developers. While GraphQL certainly has its merits, the speed at which Shopify is deprecating established technologies and forcing migrations is becoming increasingly challenging for those of us trying to build stable, long-term solutions on their platform.

It seems like every few months there's a new announcement about some API or feature being deprecated, replaced, or fundamentally altered. From the retirement of the Ruby Script Editor to the overhaul of checkout customizations, from the introduction of Shopify Functions (with Rust, no less) to the constant evolution of APIs, it's becoming difficult to keep up without dedicating significant resources to simply maintaining existing functionality.

This latest move to make GraphQL mandatory for new apps by 2025 is just another step in a pattern that's becoming all too familiar. While I understand the desire to push forward with newer technologies, there's something to be said for stability and backwards compatibility. The lack of a proper LTS strategy for their GraphQL APIs means that what works today might be completely obsolete in just a couple of years, which is a tough pill to swallow for businesses trying to plan for the long term.

Add to this the recent price hikes, the restrictions on existing features like reducing the number of markets across plans, and the constant pressure to adapt to new paradigms, and it starts to feel like Shopify is prioritizing rapid change over the needs of its developer ecosystem. It's exhausting to constantly be playing catch-up, especially when these changes often don't bring immediate tangible benefits to our end users.

Don't get me wrong, I appreciate Shopify's drive for innovation. But at some point, one has to wonder if the pace of change is actually hindering rather than helping the platform's growth and stability. It would be refreshing to see a more measured approach that balances forward momentum with the practical realities of maintaining complex e-commerce systems.

IMHO Claude 3 output is less corporate bullshit speak and more to the point. I prefer it over GPT-4. I feel like an adult when talking to Claude. GPT-4 tends to go off on a tangent quite often. I feel like a teenager stuck in a moronic conversation sometimes. I would also regularly run both side by side - in long conversations, I'll mix messages from both. Claude seems pretty good at staying on point and produces more concise output 90% of the time. My 2c

Amylyx Stock Wipes Out Nearly All Of Its Value, Crashing 82%, After Pivotal ALS Drug Failure[0]

[0]: https://finance.yahoo.com/m/b15fb91a-1e90-3259-844e-cfdb2563...

Heavy user of `z` for many years... that is until it corrupted its own database one final time. There's nothing more frustrating than a dropped or corrupted directory database just as you've got the damn thing to remember all your favourite spots on the disk.

These days I use https://github.com/gsamokovarov/jump which I've mapped to `z`. Happy days.

I run this[0] on Google Colab. The way I have it set up is to encode the meeting minutes to .ogg, push them to Google Drive, then adjust the script to tell it how many speakers there were and the topic of conversation. The `initial_prompt` really helps the model especially if you are talking about brand names, etc. that it may not know how to correctly transcribe. I've added a comment at the bottom of the Gist with some of the prompts I've used in the past. I've successfully managed to produce reports on week-long meetings (~18 hours) that were essential to get the team up to speed.

As a company we are currently shifting to Otter.ai[1] which gives good enough results for everyday meetings.

[0]: https://gist.github.com/StanAngeloff/91480fac18a74d8aff3e4cf... [1]: https://otter.ai/

I personally use Whisper to transcribe painfully long meetings (2+ hours). The transcripts are then segmented and, you guessed it, entered right into GPT-4 for clean up, summarisation, minutes, etc. So in a sense it's a great way to get more people to use their other products?

This started off as a CDN issue and seems to have spread throughout their infrastructure. Shopify CLI deployments are also affected, so is theme development. I'm also unable to access /admin or admin.shopify.com on more than a dozen stores.

This reminds me of Hiber[1]. Hiber launched in 2016 with the mission of connecting the 90% of the world that didn’t have access to connectivity. They were working on HiberBand, a protocol to communicate ultra-low power for remote IoT solutions. It sounded very promising and futuristic, a bit like a mix between Starlink and Swarm of today. At some point Hiber pivoted away from a connectivity-only business model. You could no longer use them for general purpose IoT connectivity, instead they focused on building end-to-end solutions. Hiber today markets itself as an "off-grid asset monitoring" solution focusing heavily on the oil industry.

I suspect there is some underlying issue with these high-latency / low-bandwidth solutions that just makes them impractical in real life.

[1]: https://hiber.global/hiberhilo/

I've been through something very similar with Mailgun. Mailgun was acquired by Rackspace in 2012. «Rackspace will make Mailgun available to its Rackspace Hosting customers for integrating cloud-based email services into applications and websites» Up until the acquisition, Mailgun was a solid product with a core set of features, i.e., an «API for creating and managing online email inboxes for apps and websites». And then it grew, and grew and grew. In my own view it tried to do too much which inevitably meant pricing changes for its customers.

SendGrid started off similarly, very early on it was named smtpapi.com. The focus was on the API, developers, UX, etc. I keep getting reach out emails from Twilio these days trying to upsell me on all the amazing "new" SendGrid features they have in 2022 (of which I need exactly none).

I don't have high hopes for Postmark's acquisition. ActiveCampaign will start "integrating", then comes the raft of new features, products, solutions. Today's Postmark does very little of what the competition tries to offer. At some point I'm already preparing myself for the inevitable price change ^^.

All in all, good pay day for the founders after 12 years of bootstrapping, not much to celebrate as a customer.

> Improved navigability sounds useful.

It does, indeed. It does also remind me of GitLab and their quest for finding the best navigation [1]. It was every few versions that the nav kept changing... from horizontal to vertical to fly out to top/side, etc. Looked like bikeshedding on the outside as it seemed to be easier to iterate over the design of the product rather than the actual product. Ultimately what their UX team settled on has been pretty stable for a while.

[1] https://about.gitlab.com/blog/2019/07/31/navigation-state-of...

> In mid-2021 we started to think about modernizing MDN’s design [..] with an emphasis on improved navigability and a universal look and feel across all our pages. > > A new logo, chosen by our community [..] We worked closely with branding specialist Luc Doucedame

Wait, this sounds like redesign for redesign's sake?

> Coming soon: MDN Plus [..] In the coming months, we’ll be expanding MDN to include a premium subscription service based on the feedback we received from web developers > Support MDN and make it your own. For just $5 a month.

Ah, this makes sense now. It looks like MDN is trying to rebrand and reposition itself as a subscription service going forward.

I remain optimistic, it's been a great resource over the years.

2022 Checklist for Confirming a Service is Down:

[ ] Do a Twitter search [ ] Scan through HN [ ] Check with peers [ ] Consult downdetector.com

…in this order. Do not bother with official status pages.

See https://stackoverflow.com/questions/3224878/what-is-the-purp...

    $ type :
    : is a shell builtin

We've moved from AWS to Cloudflare and I can confirm Workers are awesome. The only downside is lots of work required to get npm packages running properly. Due to the lack of Node.js' runtime, bundling, shims and stubbing required to get stuff like "fs" and "net" compiling. AWS is more mature, has a better community support around it and generally the tooling is more stable. Workers are quickly catching up, though, with great new additions such as ES modules, custom builds, Miniflare, etc.

The first time $350 was hit Reddit investors were still building up momentum - it wasn't a bad play given all the money pouring in. I personally executed several trades at and around $350 and all were profitable. The frenzy eventually died down and at that point there was no turning back. I feel sad for all my friends stuck holding shares they've bought hoping for the MOON!

Some years ago, likely around 2010, I had the priviledge to visit Culham Centre for Fusion Energy (it was called UKAEA Culham back then). During my visit, we were shown around various experiments and after a long evening, we had the opportunity to attend a Q&A with the staff. The question everyone wanted to know the answer to was When will it be ready? The scientists and everyone else involved honestly thought 10 years would be more than enough time for the technology to mature and find its way into everyday lives.

Commercial fusion energry, always 5-10 years away.

I don't usually rely on a specific version and am fine running a release (or five) behind. If you don't need the latest and greatest pointing to an archived old release may save you some hassle.

I rely on those static builds and they are a lifesaver. Careful with linking to this website from automation scripts, etc. as the downloads are regularly swapped and only certain versions (IIRC latest in a point release) are kept available. E.g., 4.3.0 will be taken down once 4.3.1 is available, however 4.3.1 might be permanently moved to the archive if it shall be the last 4.3.? release before 4.4.0.

Excellent series of posts. I've been following along with no prior Rust or OS programming knowledge and have found it immensely refreshing.

Looks to be {relatively} low as the average daily payout according to https://www.sebenz.ai/company/about is $5.85

Ah, Komodo! I wrote a bunch of extensions for it [1][2][3] that got me started with XUL. I loved both. XUL always felt like every feature you can think of was available and thoroughly thought out, you just had to discover it (not always straightforward given how lacking the documentation was, especially for internal/private APIs). XUL was way ahead of its time.

[1] https://github.com/StanAngeloff/komodo-html-toolkit [2] https://github.com/StanAngeloff/komodo-fuzzy-open [3] https://github.com/StanAngeloff/komodo-aero-theme

Over the past 15 years, I've had the opportunity to work with more than a dozen MLM companies in the UK. This won't be a popular opinion, however I believe great quality products and a self-regulating industry can create businesses that are far from being a Ponzi scheme but are legitimate companies with ethics, great support and opportunity to earn decent money.

I, oddly, see a different pricing structure on that page. Full-time development Standard €55.10. This is assuming 100 active hours. "Active" being defined as "a user is connected […] via the browser-based editor". If I log in at 9AM and log out at 5PM every day for an average of 20 working days each month, I'd probably run ~€90 a month on the standard instance (and considerably more on the Premium instance). My current laptop cost around €1, 200 and I can easily provision a new one at any time using an Ansible playbook[1].

I am most definitely not the target audience of VS Online, which is perfectly fine. Someone out there needs this for it to have been built by MS.

[1]: https://github.com/StanAngeloff/longitude

I want to like this, but having no support for Firefox, no SLA during the preview[1] & being unable to figure out the pricing model[2] makes that very difficult.

[1] How safe are my projects and what guarantees do I get that my work won't be lost?

[2] For cloud-hosted environments, each environment instance is billed based on the number of consumed “environment units”, which are calculated according to an environment’s instance size, the total time the environment is active (i.e. a user is connected to it via the browser-based editor or via a client such as Visual Studio Code) and the total lifetime of the environment (base units).