Posterous: A reminder of why you should own your online presence | Hacker News Reader

Posterous: A reminder of why you should own your online presence

dendory.net

46 pointsdendory26 comments

Open articleView on HN

Sure... but I want to spend less than 15 minutes a month administering said online presence.

That means not looking to see if there's a 0 day exploit.

That means not even thinking about upgrades.

That means not having to be the one to figure stuff out after an upgrade goes wrong.

That means someone else worrying about if it's being overloaded. (The DDOS of love...)

That means being able to quickly send off an email (a la posterous) or similar to post. Markdown is preferable.

That means, unlike static generators, I don't have to think about re-deploying when I do anything.

I have too much to do with programming and administrations for pay -- a blog is not where I want to have to think after hours. Administration isn't fun anymore. I want to own my service, but I don't want to have to babysit it.

I don't even mind paying for the privilege, though the amount needed to justify it might be more than I'm willing to pay.

What is everyone else doing that fits the above? This seems like dumb duplicated effort.

This is why I use GitHub pages. Jekyll is good enough, and with GH pages it's super simple. If GitHub were to ever do something weird with it, I can get any static hosting and do the jekyll stuff myself and still control it.

And a static site is much simpler than, say, a Rails-based site (the other thing I was considering when setting it up). Any hosting will do, and I only have to worry about exploits in the web server, not the framework.

That's the route I think of going, but I still don't think that's controlling your presence on the web -- you're losing all your google juice if you ever have to move.

What's the difference between that and any other system that allows you to export your data?

You can use your own domain with GitHub Pages, then you don't lose any 'Google juice' if you move to another hosting provider.

You should look into Squarespace. They built their own platform and do all the maintenance and upgrades. I've never heard of a Squarespace exploit or of a Squarespace site going down because it got too much traffic. They did almost go down in the hurricane, but supposedly they're now working on failover plans to ensure something like that doesn't happen again.

I've never heard of a Squarespace anything because I've never heard of Squarespace, why is this a reason to think that it is more reliable than alternatives?

Because they charge for services and are (as far as I know) profitable.

I should prefer something I never heard of because it charges? None of the alternatives charge for services?

Free services get shutdown regularly unless a benefactor with deep pockets can be found, simple as that.

http://en.wikipedia.org/wiki/Squarespace

It's a decade old, has 100+ employees, advertises on TWIT and TWIST. That you've never heard of it implies that you've been busy or uninterested, not that SS is obscure.

Life has no guarantees if that's what you're looking for. ;)

You're a programmer and apparently know systems, why does your service require babysitting?

Rhetorical, but yes, as you imply, convenience is a luxury and has a price. What other people are doing is either babysitting or paying more for the luxury of not babysitting.

Because 15 minutes a month quickly gets eaten up by writing my own monitoring scripts, and because, frankly, I don't want to. I find it boring. I do enough administrative tasks during the day to bring tedium back home. I find it an impediment to actually creating.

That's the appeal of something like posterous, even if I don't "own my presence."

>Own your domain name. If you want to post content, anything more substantial than cat pictures, things you may want to come back to in the future, you should own your domain name. It's cheap and ensures that your links won't have to change even if you need to move to a new platform.

I have been meaning to get a set of personal (as opposed to project-centered) domains for a while for exactly those reasons. Should I get the .me and the .name along with the .com? What best practices would HN recommend?

Buy just one. Any more and all you're paying for is the utter retardedness behind TLDs and domain name system. What are you going to do with the .me and .name? Just redirect them? If someone else comes along and you've not built the domain authority you'll still lose out if they build a better online rep than you, it really doesn't matter if they get the .name or the .org or the .eu or the .xxx or the .arenttldsgreatandnotatotallystupididea.

Preferably not from a really dodgy country.

I'd avoid .xxx, some web filters seem to block that indiscriminately.

.com only.

Forget .me and .name. No need for those.

(I've been a registrar since the ICANN start and in domains since 1996)

All taken.

.im?

Failing that, drop over to http://domai.nr/ and try to find a domain hack based on your name. Once in a blue moon, people get truly awesome hacks. http://rog.ie/ springs to mind.

The number of small businesses that have abandoned their web sites and are relying on Facebook as their web presence is frightening.

I already mentioned this on another thread today (https://news.ycombinator.com/item?id=5229643), but this is what we're trying to solve with Tent (https://tent.io).

This is very interesting. Can I ask some questions? How is what you are doing different than diaspora? Is it interoperable with it? And what kind of traction have you gotten?

Tent is just a protocol that can be used for all sorts of applications. It's easy to make microblogging or traditional social networking apps (think Twitter and Facebook or Diaspora) on top of Tent, but apps like Dropbox, Evernote, Tumblr, Loopt, etc are also possible.

Diaspora is an app with a great UI and its own (unfinished) protocol. We've been working with the Diaspora* team (http://wiki.diaspora-project.org/wiki/Diaspora,_Powered_by_T...) to create a port of the Diaspora* UI to Tent (basically take the Diaspora* front end and a standard Tent app back end). That would let the Diaspora* community continue to use the app they love, but take advantage of the other capabilities of Tent and apps in the Tent ecosystem. It's easy to import the old posts from Diaspora as well. The only problem is that the Diaspora* community is currently based on a pod infrastructure (dozens of medium community-run servers (100s-1000s of users/server). Tent assumes that users will either run their own server or use a hosting provider (think wordpress). So far, the only Tent server implementations capable of hosting thousands of users are closed source and belong to the hosting companies. Eventually that may change, but it means there's no drop-in replacement for Diaspora* pod admins. Users would need to either set up their own Tent server (really easy to push to Heroku https://github.com/tent/tentd-admin) or use a hosting provider. There aren't any UX or technical complications, the D* community is just used to community run pods. As it stands now they'd need to write a multi-tennant Tent server implementation themselves. Additionally, the idea of part time admins and a giant shared server might work for apps like Diaspora (or even some Wordpress installations), but Tent is intended for highly sensitive personal data as well. I'm not sure that part time administrators should be responsible (technically or legally) for sensitive data (imagine storing your medical and financial records, persistent location data, file backups, etc on a server with a single part-time admin).

Right now there are a few dozen Tent apps of various types and a handful of server implementations. Some of the reference apps and servers as well as the protocol validator (in development) are available at https://github.com/tent/. The first Tent host currently has about 19,000 registered users. A dozen or so users are hosting their own Tent servers (similar to running your own Wordpress) and three other Tent hosts are preparing to launch. There's a lot of big announcements coming in the next few months that should cause an increase in those numbers.

The problem isn't necessarily with services providing Open APIs/Data Exports, it's with people themselves and the trust they ensure to the services they use regularly. Pretty much all of this advice most of us should know: along with regular backups and different passwords/password manager for different sites; and I am willing to bet a large proportion of people will know this advice, advise it to others, but not follow it themselves. That is until its too late.

Even when following said advice it is not fool proof. So you own your own domain name. But do you really own it? Not really, you are essentially leasing it from a registrar who probably have terms and conditions for seizure of that domain name for whatever reason.

This is why you shouldn't use GeoCities or MySpace. </sarcasm> Just because a single website is shutting down doesn't mean you should spend extra time installing stuff on your own. I disagree with the article.

Also see:

http://indiewebcamp.com/why

http://indiewebcamp.com/POSSE

I agree with the sentiment. I just wanted to point out that you could own your own domain and point it at posterous. Not perfect, but still.

File this in the category of: making backups, not using GoDaddy and using a password manager. It's easy, everyone knows it, but yet most people don't take the tiny up front effort to do it right.

Just last week, another GoDaddy rant. How quickly are these startups closing? How many startups' products do you use that have no business model other than "attract users and content and figure it out later". Enjoy when the next one that you're "actually using" succumbs to the same fate!

Reply on news.ycombinator.com