I tried Zed for a few days and loved it, but had to switch back to Cursor because I've become dependent on Cursor Tab.
The problem is that Cursor Tab seems kinda psychic, and I didn't realise how conditioned I'd become to just expressing a few keystrokes in the right place and have Tab pick up my intentions. If you're refactoring, you can move between files and it'll remember what you just did in a tab you just closed, and work out how to make the same changes here.
It's also really good at picking up patterns and the right imports from the whole repo. It seems to be working with a much larger, more persistent context.
I tried Zed AI, Copilot, and Mercury. All three seemed forgetful after a year of Cursor Tab. I wish there was a fix because literally everything else about Zed was an improvement.
Thanks for the reply! I gave it a solid week at the start of this month, about 3 weeks ago.
The only other thing I didn't like (and this is subtle, but I had not realised that I took it for granted): In Cursor, you can close and reopen a tab, and the undo history is kept. Working on a large repo, this comes up surprisingly often.
Save -> Close -> Assume everything is fine -> Do something more -> Oh no! That was a mistake! -> Cmd+Shift+T -> Undo. Cursor will comply; Zed says there's nothing to undo. So you're back to reverting chunks of code in git.
I will try Zeta again shortly. The biggest thing I'm looking for is that I can move between files and the context of recent changes persists, and the assistant treats it has highly relevant. Comes up a lot when refactoring.
That theory is based on art springing up in locations where Massive Attack are playing. It's a bit odd because by that logic it could also be any of the dozens of people needed on tour.
We spent our first year as a startup hacking on laptops in the attic of Smithfields, along with a dozen other startups. No idea if Innovation Warehouse is still up there[1]. If you arrived at work early enough they'd still be hosing the blood off the tarmac from the early morning market.
Most startups moved out to WeWork as soon as they could turn a profit. But hey, it was cheap office space in super-central London.
I have many happy memories of weeks at Herrang. In fact I got offered my first job in open source there after a chance meeting, and my first startup can trace its origin to some conversations with other hackers in between dance classes.
Let's be clear - this affects client-side apps which include the template compiler (ie. the full vue.js bundle). Right? So apps built with Webpack, and using vue.runtime.min.js should be safe?
I'd imagine that represents the majority of Vue apps in production, but I might be wrong.
Virtualisaton is the basic task you're doing, you're making a "virtual" computer. The idea of the "cloud" is more that these virtual machines are hosted in a room somewhere far away with other virtual computers there as well.
For instance, I can install VMWare and get virtualisation going on my laptop, but I wouldn't consider that the cloud. Similar, you can get bare metal computers on the cloud which aren't virtualised.
Offtopic, but does anybody know how the diagrams in this article were generated? I'd love a simple piece of software to generate beautiful, straightforward pictures like this to explain architectural problems.
The problem is that Cursor Tab seems kinda psychic, and I didn't realise how conditioned I'd become to just expressing a few keystrokes in the right place and have Tab pick up my intentions. If you're refactoring, you can move between files and it'll remember what you just did in a tab you just closed, and work out how to make the same changes here.
It's also really good at picking up patterns and the right imports from the whole repo. It seems to be working with a much larger, more persistent context.
I tried Zed AI, Copilot, and Mercury. All three seemed forgetful after a year of Cursor Tab. I wish there was a fix because literally everything else about Zed was an improvement.