Reminder for everyone that is using Azure TSI and is working towards the original deprecation date of 31 March 2025 [1], you should know that this has been accelerated to 7 July 2024 [2].
At my current company we are close to finishing our migration, so this has negligible impact for us.
P.S. If someone from Microsoft leadership is reading this, pushing this deprecation notice on a Friday, that only ~5 weeks remain instead of ~10 months, feels a bit wrong. Although the original deprecation notice was over a year ago, so thank you for that.
Have you tried Tailscale? I have and our company is going to migrate in the next year or so because it does a LOT of stuff better.
As for your rule limiting submissions to the front page to only one every month. Who’s gonna moderate that? Are blog posts about the product also prohibited? Cause that is how I learned about Tailscale in the first place.
My point being, no I wouldn’t want a rule to moderate something like this. Because I come to hackernews to learn about these types of stuff.
It's because they're favored by the HN community. Few startups manage to achieve this 'darling' status on HN.
I wrote about this phenomenon here: https://news.ycombinator.com/item?id=30070287 - in the bottom third of the comment, starting with "Stripe succeeded on HN". (Skip the tedious stuff before that.)
I used ZeroTier in the past. I have migrated to Tailscale because:
* It allows for faster transfer speeds.
* It's less laggy.
* It takes less time to negotiate the P2P routes on "difficult" topologies (both peers behind CGNAT, for instance)
* It actually just works (with ZT it was hit-and-miss. mostly hit but very frustrating when miss)
* It was easier to run on my router and just use that as a gateway, as opposed to meshing all my devices (I did this with ZT but there was quite a bit of configuration and trial-and-error required -- TS just works)
Perhaps TS gets more coverage because people like it better?
Disclaimer: not affiliated with either company, just a happy user - despite the issue with identity providers.
Besides Tailscale not being a YC company, YC companies do not get more upvotes/weight/whatever on HN. They have to fight for attention alongside everybody else and I can tell you that they mostly find it just as frustrating.
There is one big exception to this, which is Launch HNs as described in https://news.ycombinator.com/newsfaq.html#yc. Each YC startup only gets one of those, and they're clearly indicated.
ZeroTier was arguably the pioneer in this market, but aside from being too early, their product was not nearly as polished or fast. Their clients are also closed source and they actively went after third party coordination servers, wherenas I hear headscale mentioned by tailscale employees all the time.
Tailscale goes for adoption by convincing enthusiasts that might also be decision makers.
ZeroTier thought it could just survive on its own merit, in a time where "Zero Trust" as a concept was almost entirely unknown.
> they actively went after third party coordination servers, wherenas I hear headscale mentioned by tailscale employees all the time.
ZeroTier founder here. We never did anything remotely like that, so I'm curious about where in the world you heard that. Is someone out there saying that?
We regularly tell people about FOSS stuff like this:
We did adopt an AGPL or SSPL style licensing strategy a while back to discourage closed for-profit attempts to basically expropriate the product. We are considering going back to a more liberal license in the future since that issue seems to have lessened. But we never had any issue with FOSS or not-for-profit stuff using ZeroTier.
No i haven't tried it since it's not something I need. However my point was not about Tailscale being a good or bad product. My point was about the amount of a product related articles in the front page of HN.
Let's suppose I had a product in the same space as Tailscale. How was I supposed to get any customers if everybody in HN knew about Tailscale?
About the rule: I think once a month is enough; the article would get traction and you d still see it and learn about it. The duration is not really important, it could be 15 days is a month is too long. But definitely not every 3 days like it is now!
Finally, the moderators would do the moderation. Users would flag frequent Tailscale articles and the moderators would demote them.
At my current company we are close to finishing our migration, so this has negligible impact for us.
P.S. If someone from Microsoft leadership is reading this, pushing this deprecation notice on a Friday, that only ~5 weeks remain instead of ~10 months, feels a bit wrong. Although the original deprecation notice was over a year ago, so thank you for that.
[1] https://azure.microsoft.com/en-us/updates/we-ll-retire-azure... [2] https://azure.microsoft.com/en-us/updates/we-re-retiring-azu...
[EDIT] Links were not working.