I’ve tried GSD several times. I actually like the verbosity and it’s a simple chore for Claude to refresh project docs from GSD planning docs.
Like most spec driven development tools, GSD works well for greenfield or first few rounds of “compound engineering.” However, like all others, the project gets too big and GSD can’t manage to deliver working code reliably.
Agents working GSD plans will start leaving orphans all over, it won’t wire them up properly because verification stages use simple lexical tools to search code for implementation facts. I tried giving GSD some ast aware tools but good luck getting Claude to reliably use them.
Ultimately I put GSD back on the shelf and developed my own “property graph” based planner that is closer to Claude “plan mode” but the design SOT is structured properties and not markdown. My system will generate docs from the graph as user docs. Agents only get tasked as my “graph” closes nodes and re-sorts around invariants, then agents are tasked directly.
claude code (CC) plan mode: In a normal CC window hit "shift-tab" until you see "plan mode on" in the lower left hand corner of the TUI.
Now all you really have to do is chat with claude about what you're thinking about building.
In plan mode, claude can't edit anything and has some extra "you're an expert at planning!" prompts prepended to your initial message in plan mode.
And then either when you're ready or Claude thinks the "plan" is gelling, it'll suggest stopping and let it write up a detailed plan. CC will dispatch some "planning agents" with prompts that your 'main' CC has crafted for that agent to plan for within the context of your conversation and what parts of the codebase it should look to integrate/explore.
Once all that is done, it will display it to you and then offer to "clear context and implement" - where it will just get to work. Or it will offer to go back to chatting and resolve whatever was misunderstood, or if you had a new idea you wanted to mix in.
These plans are saved as markdown in your .claude/plans directory.
Plan mode is handy on the on-off. But if you enter another plan mode, thinking claude would learn from, or build off a previous plan spec, it won't unless you explicitly say something like "read previous plan <path to plan file> and re-use the scaffolding directives for this new project"
Your employees are there to make your life easier? As their manager? Do you demand they make you coffee? Rub your feet?
I've been doing this for decades, and I've never seen that attitude work with any 'leader.' I'd hate to work for you. Ever hear of servant leadership? Or hear the line "My job is to clear the runway for you"
Managers are cost centers, 'your' employees are what keep you employed, give them the respect they deserve.
That's not at all what I mean. What I mean is that I am responsible for the output of my team. If someone I am supervising does a bad job, is hard to communicate with, etc. it means that one way or another I have to do more work, which reduces the total output of my team. It can also lead to inferior service, angry clients, adverse outcomes, etc.
can't sign up with gmail or "personal" email addresses? What if I want to evaluate but I am not ready to inundated with sales calls? My 'work' email domain is one that many vendors would love to see in their CRM. I always sign up with disposables first.
I guess I should thank you for saving my time? Plenty of others in this space.
I keep a similar kit of tools. However I still had problems fitting _everything_ i typically need in a small case. Typically I'd have to decide if I needed more socket or driver options and add/remove tools to accommodate.
I have since solved this problem:
1) Tool-Check PLUS from Wera: fantastic and highly condensed set of sockets, bits, ratchet and driver. all high quality.
2) Leatherman ARC: replaced a more basic multi-tool. I like the high quality blade material and the whole thing is very durable, i can squeeze and crank the pliers without worry.
The leatherman is a bit of a splurge but I highly recommend the tool-check for tool-using EDC minded folks. They make good gifts too.
I regularly play UHD HDR content I ripped from my blu-ray collection to my 4k AppleTV over 1Gbs Ethernet. I use the Infuse app instead of Plex, no stutters, no issues, looks great!
sidebar: I've had a lot of problems with the AppleTV plex app, it does not play some HDR content well, or at all. Plex on my nVidia Shield works well.
One of my greatest treasures as a young computer nerd was a bare 386(I think) chip that I received after sending away for it via an Intel ad I found in Byte magazine. I just had to cut out part of the page and mail it in. Several months later I get a package back with the naked processor glued to stiff card along with a low powered magnifying glass to scope it out with.
Sorry, im a nerd/geek, and I know god does not like nerds/geeks. Our mutual interests do not hold up in heaven. We are all basically wasting our lives on things 99% of girls don't like. And eventually you won't like it either.
I committed a few sins lol and became a noble Saint, so God started speaking to me, decided he didn't like me, and now I'm stuck with a bunch of useless knowledge.
You must not know people very well imo. Most people just want fun/money, that's all they think about. A lot of people lie, cheat, steal, hurt people.
Who do you think triggers these hormones to start being produced?
There's a lot we can do with science that affects our bodies, i.e. steroids. Your citing behavior after hormone manipulation doesn't make sense.
God works slowly on people, once he makes a decision your basically stuck with it. He would not cure balding magically or otherwise just because a person changed their character or repented and reformed.
God might work slowly, but hormone manipulation works fast and predictably. I'm not sure why you'd choose God over hormone manipulation. One is something which is proven to work and proven to work fast, one is riskier, slower-acting and you are on the whims of someone.
I would really love to see an official "hide my email" extension from Apple. I use this feature so much that I go out of my way to use it since I don't use Safari and that's the only integration for HME.
Bonus is that I use a separate subdomain of my custom domain for disposable email addresses which means it never fails email checks (some don't like disposable/temporary email domains).
Doesn’t seem like a bonus. Your custom domain could just be blocked wholesale. Hide my email uses iCloud, not a separate domain. iCloud addresses probably won’t be blocked as that’s the default email given for hundred million plus Apple accounts.
Fastmail by default uses @fastmail.com addresses for masked email, same as their cheapest plan, so when I used it there was never anything blocked except the occasional site that only accepts gmail addresses (yep, they exist). Personally I prefer it that way for personal use, the custom domain sacrifices anonymity, but it might be nice for business users.
Fastmail is still like 0.1% of users. If you're a "growth & engagement" company you're probably better off banning that domain wholesale and as a bonus will get rid of tech-savvy ad/tracker blocker users without making any impact to your target market.
The advantage of iCloud is that it's a domain laymen use - those same laymen the "growth & engagement" scum wants to track and spam. They can't just ban it wholesale without alienating a large chunk of their target market (and a pretty lucrative one at that, since Apple hardware is expensive).
A while ago I ran into the first site that told me I couldn't use my fastmail masked email: remove.bg.
I don't know if they block every fastmail.com address, or if they somehow check if it's a masked email.
Lots of random weirdness for me trying to use mine. Most recent example was trying to checkout as a guest on Little Caesar’s app. Kept declining my Apple Pay transaction without telling me why — changed the email and it worked.
There’s only been a handful of times I’ve used it, maybe 12? At least a few of those times it wouldn’t even let me submit it saying it was an “invalid email”. Couldn’t even get past the validation.
Any idea if using a custom domain would have worked better here? I wouldn't be surprised if some places only accept email addresses from a small set of "known" providers
Companies blocking disposable email domains are doing it to prevent many users from using their service with a disposable email.
A custom domain that only I use specifically for disposable emails would look indistinguishable from any other custom domain out there, and nobody else would have used it for them to even be aware of its existence.
To block it pre-emptively, they’d have to either be omniscient or block every single custom domain in existence. The former I highly doubt is the case, the latter would generally do more harm than good to them.
I used to use an MX record that had mailinator handle nospam.jrock.us email addresses... but stopped doing it for anonymity reason. ("whois nospam.jrock.us" whoops there's my home address!)
The alternative I currently use is letting Gmail handle the spam. I used to be big into jon-foo@jrock.us for "foo" and that sort of thing, but every address ended up on every spam list anyway, and the filtering didn't increase the signal to noise ratio.
For true throwaways I just use mailinator. If I want to receive email from someone someday, I can just create another account. If they spam me, Google will filter it out. So it goes.
You use jon-foo-randomchars@example.com because then you can be sure where the address was leaked from and which companies are selling your email address, and to provide additional signal to the spam filter. If you use just jon-foo@, then that's guessable and you can't go off on them for selling your email address when they said they wouldn't.
I generated a “hide my email” forwarding address specifically for my HN profile last weekend.
Not via a chrome extension, but it’s pretty easy to generate one on any Mac (maybe even iOS?) in system settings. You can name the forwarding address to have different ones for different uses
Edit: re-read your comment and sounds like you’re already doing this manually, like me. I agree it’s a hassle and would love to see a more native UX that doesn’t involve opening system settings
Shameless plug: I've built the unofficial "Hide My Email" browser extension [0], available both in Firefox [1] and Chromium [2]. Tried to make it as frictionless as the Safari UX, which proved to be a challenge given the lack of native HME APIs.
Same for Apple Pay. Given that I mostly buy thing on my computer, and that I don't use Safari there, I basically never use it, even though I would really prefer it over entering/auto-filling my card number.
Apple’s Hide-my-email service is only useful to people who don’t have their own domain-name for email. I assume most of here on HM have a vanity dot-com or dot-me that we just point to GMail (or maybe Office 365 if you lean that way) - all those services (not to mention self-hosted) allow us to set arbitrary, catch-all, and disposable addresses (even the perpetually un-cool O365 supports it now too).
I’ll never use Apple’s Hide-my-email service until they let us use it with our own domain-names. It’s my email mailbox and my dodgy account registration, not Apple’s.
I wouldn’t say it’s only useful for people without vanity domain names - it also adds a level anonymity should the user database get leaked (IE: it’ll be a random HME account that will look like every other, vs your specific domain name)
Also FWIW you can use it with custom domains - I use it with fastmail and have since day one (their version of it isn’t as tightly integrated as Apple but they offer this service as well). Hope this helps!
To confirm, you're saying that Apple will generate something like "randomString@yourDomainName.com"? If so, how can I set that up? I don't see any relevant settings on my phone's iCloud settings page.
Like most spec driven development tools, GSD works well for greenfield or first few rounds of “compound engineering.” However, like all others, the project gets too big and GSD can’t manage to deliver working code reliably.
Agents working GSD plans will start leaving orphans all over, it won’t wire them up properly because verification stages use simple lexical tools to search code for implementation facts. I tried giving GSD some ast aware tools but good luck getting Claude to reliably use them.
Ultimately I put GSD back on the shelf and developed my own “property graph” based planner that is closer to Claude “plan mode” but the design SOT is structured properties and not markdown. My system will generate docs from the graph as user docs. Agents only get tasked as my “graph” closes nodes and re-sorts around invariants, then agents are tasked directly.
reply