Six Degrees of YC Separation | Hacker News Reader

Six Degrees of YC Separation

medium.com

94 pointskmack37 comments

Probably more accurately titled "Bay Area Tech is a Small World After All"

TLDR: Chekr Director of Solutions, formerly Director of Strategic Sales (and first business hire), starts client on-boarding company (nice idea, streamline a lot to minimize hand-holding).

I find that the mechanisms that make the tech world small are interesting in themselves. Graphs are made small (in the graph-theory sense of the shortest distance between typical nodes) by adding connections.

This anecdote about how some particular connections were formed and later led to real opportunities seems like a good example of how things often happen in the Bay Area tech world.

Perhaps someday there’ll be a general theory of how opportunities are created, but in the meantime if you want to learn how things happen, there are only anecdotes to learn from.

> Hiring people that have a track record of taking projects from inception to delivery and who actively seek out projects to own is critical

Doesn't that regard the first few early hires (who'd take some form of senior/management positions later on)? For me a person who delivers projects means somebody who's very involved in the process; somebody that calls significant decisions about the product.

Should all hires be like that? Isn't being able to execute the tasks you're assigned enough?

I would go one step further and say if you have full of go getters, some might get bored when the inevitable “just execute the assigned tasks” work comes up. This is what I hear happening at Google a lot: they only hire A players, but many get bored (and eventually leave) when they are assigned B or C-level work that needs to get done.

I was visiting a large successful startup that had a "no managers" mantra--they hired engineering A players who were expected to work on the right things.

It turns out that people want to work on the fun and challenging tasks. Nobody put the effort into the billing system and they were doing a poor job of collecting money from their customers.

That's the thing about any organization, there's plenty of grunt work and "boring" tasks to go around, but they need to get done. Scaling up any organization is an exercise in making sure that the right things get done.

I once interviewed at a place that had "no managers' yet kept referring to the "MGMT" team. Made me laugh.

I’ve yet to see a company with no managers work out. Usually it signals lack of bandwidth to set up management, low perceived value, or pessimistically a cost-cutting measure to double up on responsibilities.

Valve.

Rumor has it that the environment is toxic, and that a hierarchy and tribalism formed in the vacuum of leadership and a defined culture.

https://www.reddit.com/r/valve/comments/8zmp07/former_valve_...

Perhaps true... but they are wildly successful as a company (if profits == success). There are a few other examples of flat organizations out there that have been successes, including Github (for a while).

Well, that can still work out if an A player is hired that thinks billing is an interesting problem to work on (Money! Whoo!), and in general, folks that, when given "boring" grunt work, will automate the heck out of it.

This does require attention to diversity of thought and background, but not necessarily the sort that results in B and C players being hired.

> Should all hires be like that? Isn't being able to execute the tasks you're assigned enough?

If you manage to build an org where everyone is a self-starter who takes on projects from ideation to delivery repeatedly you work in a crazy awesome env. There are some people that are really good at getting work done, but if you don't give them a clearly defined task, they will flounder.

When you are in startup land, most of what you are doing is undefined. You need someone who is really good at saying "is this working? Should I keep doing this? How do we do this better?" Over and over and over again until you find product market fit and aren't burning through cash.

If you are running an enterprise mature SaaS product a lot of those variables are solved (but possibly could be optimized). The majority of people don't need to be solving for profitability. In startupland of 10 employees or less, almost everyone should.

It's the same logic that a "startup founder" may not make the best "enterprise CEO". An "early hire" may not make the best "gear in a 300 person machine".

> You need someone who is really good at saying "is this working? Should I keep doing this? How do we do this better?"

And sometimes you need someone who is really good at saying "let's just get this done." Diversity is a good thing on any team, even in a startup.

Such or will have horrible politics as all those autonomous people fight with each other over who will call the shots and who will control non-existent resources.

There is handling uncertainty and there is organizing workplace to maximise unceirtenity. You described the latter.

I have been burnt by this attitude as an employee many times. What people usually mean when they say "I want to hire someone responsible to own the project" is "I want to hire someone who will pick up my vision of what needs to be done, will make it happen, taking the blame in potential conflicts along the way and will then let me get the credit for it".

I ended up starting my own business and I think that's the only realistic scenario for someone who likes "taking projects from inception to delivery and who actively seek out projects to own".

What exactly is meant by "commits" that you talk about in 1:1 and meetings? Is that more like "commitments" or "git commits". The former I agree, the latter I think is in many cases a terrible measure of productivity, because it discourages activities that don't directly lead to code, such as pairing with your junior, or helping a customer, or providing mentorship for others, etc. Remember you'll get what you measure.

"Commitments", not "git commits". For revenue teams, this could be signing a contract, scheduling a certain number of meetings, drafting a blog post, etc. Similar parallels for any team within a company. The key is being able to break larger projects into manageable weekly commitments that are specific and measurable.

Awesome, thank you

Did you use the word in error or is this another example of the trend of start-ups using verbs to replace nouns?

I so agree with the ridiculousness of this behavior. When people talk of an "ask" meaning "request" i cringe at the embarrassment of their need to be jargon-cool rather than individuals.

Nice Kyle!

congrats and good luck!

I thought that since this author's startup is sorta a 'reddit for politics', I'll plug my own startup (currently self-funded) which is a 'chatroulette/duolingo for politics': https://dinnertable.chat/

That was his first YC application, and it got rejected. Looks like his current startup (which is in YC) helps companies with onboarding.

Your implementation looks nice! The idea of yet _more_ people to discuss politics with however... is not for me.

I think of it more as a system to help people improve their political reasoning. Too many people are stuck in bubbles, and 'reaching across the aisle' cannot be solved by just showing the "facts"... it needs to be an emotional journey best carried out through interpersonal connections. The question is... will people be willing to have these difficult conversations? I'll just have to see with this startup. heh

Cool, seems like it'd attract exactly the people who don't need it though. ;- )

I'll sign up I think!

Edit: While I was signing up, I tried copying the verification code from the email, and it copied with the period from the email, so when I pasted it into the site it didn't work. Maybe put a zero-width space in there (U+200B) or ditch the period altogether. ;- )

Edit 2: The "start login" page didn't work the first time around. In the console I saw a 404 status on something. When I refreshed, it redirected me to the right page though.

Give that signup and sign in flow a good test, I was almost ready to call it quits after trying a couple times!

Edit 3: The countdown clock is jumping around a bit. Turns out the font for the numbers (Montserrat) is proportional, and does not support OpenType tabular number variants (in CSS you select these with: { font-variant-numeric: tabular-nums; }), so the countdown clock tends to jump each time a second passes and it is updated. Maybe nitpicky but it bothers me.

Thanks for the feedback- super appreciated! The login/reg system is in need of an overhaul and is the next task on my list. Also, ya, the font choice has made the countdown timer finicky. heh

"seems like it'd attract exactly the people who don't need it though"

I know you're joking a little, but aren't the people that hold incorrect beliefs exactly the type of people that need an emotional way to connect to others (the opposites) that are informed to resolve these differences? Often I find that people with strongly held views are able to reach small compromises via personal communication, rather than any counter facts/news. Whether this particular implementation/solution works, time will tell :)

> I know you're joking a little, but aren't the people that hold incorrect beliefs exactly the type of people that need an emotional way to connect to others (the opposites) that are informed to resolve these differences?

I agree whole-heartedly, and I've seen this happen a lot on Twitter (though clearly it is not the norm there, from what most people say). My point was more that you'll have to work pretty hard to find and attract people who actually need your platform, because being closed-minded is seemingly what got them into that mess in the first place.

> My point was more that you'll have to work pretty hard to find and attract people who actually need your platform,

Revisiting this: I've curious if you have any suggestions on how you might try to solve this problem?

Not the author but good feedback for all.

Montserrat doesn't support OpenType number variants which causes a flicker.

Thanks for the help, I'm not sure what you mean exactly by 'OpenType number variant' flicking, however I've since updated the countdown clock with a monospace font.

Update: thanks again for all the suggestions! Here's what I've done...

re Edit) removed the period, hopefully that helps

re Edit 2) fixed the 404 issue: was a rare bug

User flow) I've now added Guest Pass login which will improve user flow in areas. I'm curious what exactly are some of the flow issues you hit more specifically.

re Edit 3) I switched to a monospaced font for the timer :)

EDIT #A: I've made the "signup/login" banner button to navigate to the registration page (with option to see login) instead of the login page by default. This should a much better flow as most users clicking the button will be those registering. Thanks for the tip!

Another update: I've made the "signup/login" banner button to navigate to the registration page (with option to see login) instead of the login page by default. This should a much better flow as most users clicking the button will be those registering. Thanks for the tip!

Putting a log in requirement to even see what your app does may provide a filter on usage that you don't want.

While there may be advantages, I think you'd be hard pressed to overcome the drop off from potential users who will bounce.

Furthermore, you're not delivering a clear value proposition "JOIN THE CONVERSATION you're invited to our dinner party beta". And then you have an ambiguous call to action "start login". At no point are you giving the potential user a reason to even want to login. And then you're asking the user to overcome an email signup and confirmation flow before they even get to see what your product does. Consider the case where the product isn't what the potential user thought it was, now you've wasted their time, what good is their email signup?

Why not go for a progressive engagement flow? You can still require accounts, signup and login for continual use. However you could engage the user first, then make your email signup ask. Wouldn't that be more advantageous?

The entry page needs a little work, particularly the top banner as you pointed out. It's a challenge and a delicate balance. I tried to provide as clear of an explanation on the front-page... do you feel you learned of the value proposition by reading the rest of the page below the banner? I do like the idea of a progressive flow, but the tricky thing is reducing bots/trolls (+banned accounts) from entering back into the live debates.

Not OP, but I thought the body of your landing page did a good job explaining it. However as you suggest the top banner / header could be more impactful.

Ive recently been referring to this [0] landing page guide and one helpful principle is that, for banner header copy to be good, the reader should be able to know exactly what your product is by reading only your banner header

[0] https://www.julian.com/guide/growth/landing-pages

Just an update, couple things:

I took your feedback and added a guest login to the site to help reduce the bounce. I've also updated the banner with a better 1-liner about the platform.

So now dinnertable.chat has a simple progressive engagement flow, although it can still be polished a lot... but I'm hoping this will help.

Thanks again for all your feedback, and please do let me know if you have more! :]

Reply on news.ycombinator.com