osener

I knew plenty of office workers managing just fine using OpenOffice 10-15 years ago.

Today people are much more reliant on real-time collaboration, polished cloud and mobile experiences. Fractionalized open source software has a harder time competing with this than file based boxed software workflows of the past.

Reminds me of the headline I saw a long time ago: “50 years later, inventor of the pixel says he’s sorry that he made it square.”

https://github.com/zed-extensions/tsgo does exactly this

Former. Otherwise I would need to have multiple database instances running, each with its own data and migrations to keep in sync. Plus I would need to re-do local env vars to hook all of this up for each worktree etc. And I often want to know how these branches play together, with GitButler I can bring these branches in and out with a single “apply to workspace” click.

Somehow the GitButler workflow works great for me and it is the first VC software that made me drop Magit after a decade of daily use.

I do not use it collaboratively. I use it to continuously ship smaller things while working on bigger pieces and I constantly move independent changes around to different “lanes” to ship frequently as parts of my work mature.

With Magit I used staging area and amended commits continuously. With GitButler I “assign” files or chunks by dragging them into lanes as I am happy with the changes, and when I have a logical unit I commit it. Having this multiple staging areas has been a great workflow improvement as well.

This is cool, except I don’t want to run multiple copies of my dev stack.

GitButler does this cool thing where you can work on multiple branches at the same time, applied to the same working directory. For example you can work on the css while an agent works on the admin panel on a separate branch. It would be cool to have this with a tool like FleetCode.

A lightweight alternative to Playwright, which starts a browser instance, is using an HTML parser and DOM implementation like linkedom.

This is much cheaper to run on a server. For example: https://github.com/ozanmakes/scrapedown

Since this is marked as a departure from HTML and has it's own syntax, I think this is a good point to improve upon JSX.

- Remove curly braces from props and children

    <MyComponent name="Joe" onClick=(() => canBeInParanthesis()) />
    <div>userName</div>
    <div>"Text contents"</div>

- Add punning (like how { age: age } becomes { age } in JS)

    const age = 40;
    <MyComponent age />/
    <MyComponent active=true />  // explicit boolean attributes

Python does an exceptionally bad job. After dragging the community through a 15-year transition to Python 3 in order to "fix" Unicode, we ended up with support that's worse than in languages that simply treat strings as raw bytes.

Some other fun examples: https://gist.github.com/ozanmakes/0624e805a13d2cebedfc81ea84...

I tried Zen for a few months. I like the UX, even more than Arc in many places. But it is just too slow for me on an M1 Max. I've switched back to Orion. Even though I miss some goodies such as Firefox containers, I just need the browser to get out of the way, not use too much power and not warm up my laptop with runaway processes. Plus, tree-style tabs are great.

Ah the joys of enterprise.

Once, a team at my old job wanted to try A/B testing for the first time. After some research, they picked a SaaS tool to make it quick. Simple, right? Just sign up and go.

Of course not. First they had to go through THE PROCUREMENT! (dun dun dun dun)

Those legends weren’t about to let us limp along on a humble pro plan. They wheeled, dealed, and months later emerged triumphant: a glorious three-year enterprise contract. SLAs! Volume discounts! The whole nine yards. The art of the deal baby!

Small snag: by the time the ink dried, the original team had moved on. New PM, new priorities. No one even generated an API key during the whole time we paid for this top-tier enterprise service.

For simpler use-cases I've used both https://dataflare.app/ and https://tableplus.com/ with success. They are much quicker and lighter to start-up, browse some tables and run some queries.

Free versions are generous enough for daily use as well. For example for TablePlus "The free trial is limited to 2 opened tabs, 2 opened windows, 2 advanced filters (filters are not available on the free TablePlus Windows) at a time."

Magit has been simply essential for me for the last decade. I still switch to emacs for this even though most of my coding is in VSCode and edamagit covers simpler workflows for me.

Recently I've been trying jujutsu[0]. As awesome as Magit is, it can't fix some warts of Git like how it does conflict management. But I hate having to do version control from CLI. Most of all, I miss doing a bunch of work, going to magit-status (or edamagit), staging lines or entire files and committing parts of my work separately (like, dependency updates in a separate commit before committing api changes).

I've tried VisualJJ[1] and GG[2] but I just can't find a nice workflow with jj coming from Magit. Splitting changes feel cumbersome. What is your commit workflow with jj coming from Magit?

[0] https://github.com/jj-vcs/jj

[1] https://www.visualjj.com/

[2] https://github.com/gulbanana/gg

There are a lot of options in this thread already, but I've been using Figma to create and edit SVGs and don't have any complaints so far.

I came across this randomly when I searched "Figma" on HN, and I'm very glad I did. Tastefully done app -- wishing you all luck in the world.

It’s often debated whether the public at the time was aware of the scale of the atrocities committed, whether they were accountable, and whether they could—or should—have done something. But only now am I realizing how much a certain part of the population actually does the propagandists’ dirty work by defending and whitewashing such atrocities.

Music to my ears. I’ve been trying to get close to what this promises using Remix/React Router v7 or Hono. It’s the best approach for a solo maker/small team, but it always feels like a thousand papercuts. On the other side, there’s Vercel + Next.js — the straight-up guillotine.

These folks really seem to get it. I’ll be keeping a close eye on this one.

It matters a lot if you have specific macro goals. If you don’t want to lose muscle, it is important to eat high protein especially during a cut. And keeping fat and/or carbs low while doing this is quite difficult without accurate data.

So far I've been resisting the Rust hype train but I guess I'll ditch Duolingo and do this instead.

Zen browser is exactly this. It has a growing ecosystem of “Zen mods” and has a great Arc-like out-of-box experience.

https://zen-browser.app/

I don't understand these comments, both the product and the website is fantastic. This is the most impressive solo project I've seen in my life. I'm especially intrigued by the novel grinder design; and it is amazing that you could fit the motor in there. Top notch industrial design coupled with novel approaches. Almost in too-good-to-be-true land, but I will take your claims at face value. If this all checks out, this is an amazing set of products for the enthusiasts at home at a good price point (considering the market).

And don't get discouraged by the attacks on your web design, I think the website is excellent the way it is. Leave it to some reviewers to go into nitty gritty real world stuff; your website does a great job of showcasing the design language and what's great about the product.

I really wish this tired cliché would disappear, and I say this as someone who has emigrated from a country renowned for its cuisine.

Dutch supermarkets offer an impressive variety of products, and there’s no shortage of specialty or “ethnic” shops where you can find virtually any ingredient for any type of cooking. Major cities are brimming with restaurants serving world cuisines, and people with diverse dietary restrictions are well catered to, with a plethora of options available. Plus, Indonesian and Surinamese food can be considered "local" by this point (if you ignore the historical complexity of the topic) and are simply delicious.

While it’s true that the availability of cheap street food might not be as prominent, to say the food here is “bland” couldn’t be further from the truth.

Same as Humane Ai Pin, Apple Vision Pro etc. They’re making a bet and hoping that when the world moves on from smartphones, they’ll be _the_ platform for the next frontier.

In software we see a lot of these “re-imaginings” come and go, see Arc browser, new cloud databases and myriad other things rolling the dice and hoping that they can disrupt and monetize later.

I think this is rarer in hardware because

a. It is harder, requires more planning, inventory management, higher standards and budget

b. Higher barrier to entry means not a lot of small players taking risks. I suppose it takes a technological shift like AI or spatial computing to re-imagine hardware interaction as well.

The saddest part is, when software companies come and go all that is wasted is some VC money and Twitter rants. With hardware, we’re making more junk to fill up oceans and waste precious resources…

Here is an open source alternative to this tool: https://github.com/ozanmakes/scrapedown

How does Dioxus fare as a native GUI lib? Last time I checked the focus was on web targets and both the built in wgpu renderer and skia backend were not supported as much.

I hope it will evolve into a platform that prioritizes curation and fosters collaboration, rather than the chaotic free-for-all that defines npm currently.

For too long, npm has diverged from its original purpose as a collaborative package manager for a programming language, transforming instead into a battleground for popularity contests. The result is a landscape littered with neglected packages. Developers frequently bypass the opportunity to contribute to existing projects in favor of launching their clones, driven by the allure of glory and name recognition. While I fully support developers receiving the recognition they deserve, this approach undermines the integrity of the package ecosystem. The ideal state, where packages thrive rather than peak and fade within a mere six months, seems increasingly unattainable. All the name squatting and harmful packages are worth a whole another discussion.

I find myself nostalgic for the days of package managers that championed collective effort, building an ecosystem of functional components. Packages were often named for their functionality, leading to a landscape where one canonical package existed for each task. This approach, favoring practicality over catchy branding and endless turnover, is vastly preferable to the current state of affairs

Nice project! How was your experience building with GPUI? How does it compare to other native GUI libs, if you’ve used any?

Thanks, must be a mobile issue.

all of these seem to be Twitter-like social apps. what am I missing?

On a related note, is there a Cursor.sh equivalent for Emacs?