Why Slight Failed: A Slight Post-Mortem • Hacker News Board

Re: Why Slight Failed: A Slight Post-Mortem

#61

Post by danielmarkbruce » Wed, Oct 30, 2024, 2:39 AM UTC

tyrw wrote: post #60

danielmarkbruce wrote: post #44

Earlier quoted context omitted.

"build multi-tenancy themselves".... this is a weird thing to say. In the modern day, what product company builds single-tenancy..? Even if a customer explicitly wants it, you can make multi into single easily. The other way around is difficult.

Depends on the app and the context. A lot of apps need multi-tenant access control but not necessarily separate containers etc. In those cases, "adding" multi-tenancy means maintaining an Organizations table, a Roles table, and a table to map them to Users.

Maybe it's just different worlds, but in my world people wouldn't consider an app using isolation via containers etc "multi-tenancy". It would be a single-tenancy application. But I guess there are are hybrids these days with such granular database options in the cloud etc.

Re: Why Slight Failed: A Slight Post-Mortem

#62

Post by paraknight » Wed, Oct 30, 2024, 4:09 AM UTC

tyrw wrote: post #57

DylanSp wrote: post #30

Earlier quoted context omitted.

Question about multi-tenancy - do you recommend starting with a single-tenant approach, or are there off-the-shelf options for multi-tenancy that you'd recommend using, instead of building it from scratch?

We build a system that is multi-tenant capable but can run as single tenant until you need it. Happy to chat if helpful

Not sure how to DM on here, but I just checked your landing and wanted to ask in what ways you're better than clerk.dev?

Re: Why Slight Failed: A Slight Post-Mortem

#63

Post by mlinhares » Wed, Oct 30, 2024, 4:15 AM UTC

8note wrote: post #56

CharlieDigital wrote: post #49

Earlier quoted context omitted.

Depending on the industry, once legal and compliance get involved, you can face a number of obstacles. Regulatory compliance like HIPAA, SOC2 is a common one (most small startups I've seen can bullshit their way around this for a bit); many early stage startups won't have this in place since it requires a not-insignificant investment in time and money. Some industries like life sciences will request external/3rd part…

This sounds like a business opportunity, for selling to VCs to recommend to their startups

That is one of the reasons Y Combinator startups are so successful, they get recommended inside the circle of startups funded by them.

Re: Why Slight Failed: A Slight Post-Mortem

#64

Post by tyrw » Wed, Oct 30, 2024, 5:01 AM UTC

paraknight wrote: post #62

tyrw wrote: post #57

Earlier quoted context omitted.

We build a system that is multi-tenant capable but can run as single tenant until you need it. Happy to chat if helpful

Not sure how to DM on here, but I just checked your landing and wanted to ask in what ways you're better than clerk.dev?

Clerk is great for new builds in the react ecosystem. We focus on more complex use cases where there are multiple user types (e.g. freemium, enterprise, partner, internal), multiple applications, or both. Having everything in one system removes a lot of tech debt and helps teams move faster.

And by "happy to chat" I meant by my email in my bio (which I just added to the public part), if it'd be helpful!

Re: Why Slight Failed: A Slight Post-Mortem

#65

Post by ipnon » Wed, Oct 30, 2024, 7:30 AM UTC

eatonphil wrote: post #5

I chatted with Colman years ago when Slight was around about joining forces and I was really impressed by his character and intelligence. It didn't work out between us for unrelated reasons. And then later on they shut Slight down as described here. But failure is such a good learning experience. If he ever starts something again I would definitely recommend folks to follow/join it.

Yes I always think of this stat I read where the average successful business owner is currently working on their 7th business venture. The implication is then if you want a successful company you should speedrun 6 duds.

Re: Why Slight Failed: A Slight Post-Mortem

#66

Post by ipnon » Wed, Oct 30, 2024, 7:39 AM UTC

Corporations are really good at paying people just enough money to fix problems in a suboptimal way. Actually I think that’s what they optimize for. Past a certain size they minimize costs for the “just barely good enough” solution in a microeconomic way. So if you’re selling to corporations it’s best to get down to brass tacks. Show them proof of how spending $DOLLAR_AMOUNT with you each month is going to practically guarantee them saving `$DOLLAR_AMOUNT * (1 + $MARGIN)` on their accounting books. That’s a much higher bar than “it looks cool and makes employees happy,” but the nature of business is that no C-suite could ever turn down such an offer.

Re: Why Slight Failed: A Slight Post-Mortem

#67

Post by magicalhippo » Wed, Oct 30, 2024, 9:27 AM UTC

hkgjjgjfjfjfjf wrote: post #54

magicalhippo wrote: post #53

Earlier quoted context omitted.

Besides WorkOS being stupid expensive as far as I can see, the compliance requests we get goes far beyond that. It includes our development practices, internal security, who has physical access to stuff and so on. And it's never the same, and they usually won't do any work themselves so we have to figure out how our situation maps to their 2000 custom questions.

Or you quit excusing your half assed practices, put on your big boy pants and do the work to get ISO270001 certification

We're on our way but it's a lot to do, especially as a more mature company.

If you're a startup, it might be beneficial to study ISO27001 early on, so you avoid relying on things which are difficult under ISO27001.

Anyway, my point was that just relying on WorkOS won't help that much in answering the security questionaires.

Re: Why Slight Failed: A Slight Post-Mortem

#68

Post by trash_cat » Wed, Oct 30, 2024, 10:18 AM UTC

> "Building a solution isn’t enough – it needs an obvious path forward"

This is called implementation. You basically have people who work full time just on implementation plans and executions of it.

Re: Why Slight Failed: A Slight Post-Mortem

#69

Post by louthy » Wed, Oct 30, 2024, 9:33 PM UTC

magicalhippo wrote: post #55

louthy wrote: post #36

Earlier quoted context omitted.

GDPR & ISO27001 compliance are the important ones, but depending on the industry there maybe others (HIPAA for example). You need to hire an advisor and start writing everything down. Being able to hand over compliance documentation along with proof of an audit is absolute gold. If you don’t do this, be prepared for a mini-audit on every sale (if you get that far). Sales to governments will likely come with even more…

> If you don’t do this, be prepared for a mini-audit on every sale (if you get that far). That's the position we're in, though as an older but still growing B2B we have to do this for existing customers as well. We're in the process of getting ISO27001, meanwhile we got one guy out of 40ish almost full-time answering such questions now.

It never stops, but at least with evidence of audits, evidence of pen tests, and policy documentation, it can be a little easier!

Re: Why Slight Failed: A Slight Post-Mortem

#70

Post by louthy » Wed, Oct 30, 2024, 9:36 PM UTC

louthy wrote: post #36

thenobsta wrote: post #27

Earlier quoted context omitted.

I'd love to hear more insights about on this. I'm just kicking off a B2B SaaS, have a rough idea of the checklist in my head, and am trying to balance core tech development with box checking.

GDPR & ISO27001 compliance are the important ones, but depending on the industry there maybe others (HIPAA for example). You need to hire an advisor and start writing everything down. Being able to hand over compliance documentation along with proof of an audit is absolute gold. If you don’t do this, be prepared for a mini-audit on every sale (if you get that far). Sales to governments will likely come with even more…

Edit: meant to say self-host/single-tenant in the last paragraph.