Comment by cableshaft | Hacker News Reader

Comment by cableshaft | Hacker News Reader

Yep. The way social sites encourage you to share your content via their websites and only their websites is not good for longevity of content. It will disappear over time.

I do think you need your content in a share friendly way in order to get any visibility anymore, but you also need to keep a 'non-social' version of it (linked from the shared version) if you want it to stick around for the long haul and maintain link integrity.

I still pay for web hosting, and if I ever get a solid weekend again I intend to put up a much simpler, almost early WWW homepage, and Archive.org friendly, to store my projects and thoughts on, that I plan to keep up as long as I'm still kicking (the domain might eventually change, I haven't settled on that yet).

I used to want to show off my web developer skills and make fancy pages, but that just makes me never get around to finishing and/or updating them. There's a reason busy people often have bare-bones websites, and you can still make a simple website look decent with HTML and CSS.

This is what http://indieweb.org/ has been trying to tell for a long while: http://indieweb.org/POSSE or http://indieweb.org/PESOS.

That's pretty cool that what I decided upon independently over time and with lots of different approaches already has a movement with a published manifesto behind it.

Makes me even more motivated to do this now. Thanks for sharing that.

I noticed that I post quite different content on each platform. It seems that this would be a major downside of POSSE: I don't want everything aggregated on a personal website, but rather where it belongs.

It belongs on your site; it's your content. You don't have to show it, but make sure you own it.

> I used to want to show off my web developer skills and make fancy pages, but that just makes me never get around to finishing and/or updating them. There's a reason busy people often have bare-bones websites, and you can still make a simple website look decent with HTML and CSS.

100% agree.

As a UX designer (and only a hobbyist developer), I decided the way to make my personal site easy to read was to use simple, semantic, classless HTML with a super basic responsive stylesheet. Just the good, old classic document layout for me, nothing fancy. I actually use a minimally modified version of this stylesheet: https://github.com/programble/writ

Thanks for the pointer, this is a great stylesheet to use with Markdown!

It does work well indeed with Markdown. I build my site using Jekyll and it comes out just fine.

Nice, if you want to grab something a little more complete or find some inspiration, I remember seeing something like this previously:

https://github.com/mblode/marx

I quit using Facebook around 4 years ago and have since then learned AWS pretty well. Last summer after my son turned 1, I realized I was at risk of losing all my photos because they were sitting around on random devices and hard drives. I set up my own 250GB EBS volume and went through everything I had and archived it in that volume chronologically.

   /mnt/raw/photos/2014/06/IMG_5930.JPG

etc.

I've even built tooling around it to help sync my photos from my computer whenever I plug my phone in. I also wrote a custom web server in Go that makes it easy for me to arrange them into albums and email its URL to friends and family. I even hope to eventually learn Swift and just build an app like iCloud that will upload everything directly from the phone, with no laptop required.

It's all completely overkill, but I enjoy knowing that I have full control and ownership over it. It's a nice little "fuck you" to Facebook. It's also been a fun project to work on. It's the 2016 equivalent of my parents' bookshelf full of dusty photo albums from the 90's. At least there's a decent chance it will survive the next 20 years.

Sounded all good until I read this: "email them to friends and family."

You have a weak link there. In today's day and age, most people expect not to send or receive emails. Be that family photos or birthday wishes. The expectations is to share in a pool of social media, for everyone to see (or not see).

I am not a naysayer of your method (I prefer email), but it is practically too much work to select and send emails (takes a lot of thinking and decision making). This is why one-click sharing on social media, has been able to evolve to a point where it has set such an expectations across common denominator crowd.

Not to mention the 25mb limit. Maybe he just emails the direct links?

Yes, I just email a URL to a page.

If the choice is between data security and a few family members being unwilling to click an email link, I'd choose data security.

Let's face it: the kind of people who only look at your photos when they're very convenient to look at aren't interested to begin with. Nothing of value is being lost.

That's exactly my thinking. The people I email generally open the link, look through the photos, and sometimes even write an email in response! It's the damnedest thing.

I put effort into arranging everything, and everyone else puts effort into looking at it. It's a pretty fair trade.

Somehow my mother-in-law still finds it worth her time to open my emails and text me saying "thanks"

Yes, of course. That's very true. But there's a reason I used the words "common denominator" to explain the situation. I'd also appreciate and thank receiving the email. The social media sharing makes one feel less-valued. I'd rather people individually contact, to show that they care. Social media sharing is a detached sharing. It is a lazy man's ideal situation.

Social media is giving up quality for quantity. I decided it was unhealthy a while ago, and continue to feel pretty good about that :-)

Apparently, there is more money (business?) in quantity, then there is in quality.

> In today's day and age, most people expect not to send or receive emails

People who don't care about your shit expect not to send or receive emails, just easy to digest posts in an infinite timeline while they're on the toilet.

But people who care about what you have to send them (e.g. your grandma when you send her pictures of you and your boyfriend on vacation, your dad when you send him pictures of his grandchildren, a close friend when you tell her about your new job) are very happy to receive what you have to send to them and reply to it, no matter the medium.

Yes, if you want to address 99% of your social network email is probably not as good as Facebook, but if you only care about the 10% that's closest to you, the technology doesn't matter. One of my best friends from high school still sends me paper letters, and I am ecstatic to receive them when I do.

I agree with you. I clarified similar position in a followup[0]. However, let's not just confine ourselves to grandma and personal pictures. The issue is all encompassing content; some of lesser value, and some greater. You may think all of your shared content is of value, but the receiver may not feel the same. You have to judge (think) before you direct your message. It is of course an easy choice to send baby's pictures to the grandma, but when you have to target different pictures, you have to think about the target audience. Going one level deep, you may have a story to share instead of a picture. You are now hard pressed to email because you know others may have already seen it, or something else.

Ultimately, the evolution of the web to the social media has happened due to human's lazy nature. We are not capable of directing messages of mass nature to one-on-one. We can direct messages of personal nature to one-on-one, but that's about it. For this reason, social media - versus email - has been able to make inroads into people's lives. Certainly, we feel that one-to-many is not the right approach, hence we form this or that group, to confine ourselves into self-created boundaries (groups have its own pros and cons).

[0] https://news.ycombinator.com/item?id=12345534

If you want to go a small step further you might consider using exiftool to rename all your photos by date. I do this to make sorting and identifying them easier. If you assume fewer than 999 photos/day, this command will rename JPEG files to YYYY-MM-DD-NNN (where 'NNN' = 001-999 in strictly chronological order). Run it in any parent directory when you want all photos there + all subdirectories renamed accordingly.

  exiftool -fileOrder DateTimeOriginal -recurse -extension jpg -extension jpeg -ignoreMinorErrors '-FileName<CreateDate' -d %Y-%m-%d%%-.3nc.%%e .

Make sure those camera clocks are set correctly! Just like with servers, it's easiest if you always use GMT/UTC time.

This is especially cool when you have a collection of photos from an event taken by multiple people, because you see everything in the order in which they were takes regardless of which camera took them (again, assumes clock was correct).

Yeah, I use EXIF data to sort them into year/month directories, and just keep the incremental integer IDs assigned by my phone. Seems to work pretty well. If I had two cameras at once, I'd have to start renaming them.

Thanks for the pointer. I've never actually used exiftool before - just use a go library for parsing the same thing.

I'm an exiftool user and photographer that's been using exif datetime to rename files to YYYYMMDD_hhmmss_nn.ext since 2001. I saw it as a great acknowledgement when Google switched to this format for their Camera files in Android. The only thing that differs between my scheme and theirs is the _nn. It normally says _00, but if there's more than one image taken for a second it will go upwards(_01, _02 etc).

I've had this same scheme too since 1998 using bash scripts to rename. Including the two-digit numbering for n pictures per second. Might have to change it to more digits soon. Typical for Google that they left off that refinement.

That pretty awesome.

Im just using Arq for encrypted backup to Google Cloud Storage and separately uploading free res versions to Google Photos for access since it's free. Google Photos is kinda neat because it applies face recognition and deep learning so I can search for people, places, and things.

I'd love to check out your setup if you ever make it available

I have it publicly on Github, but nothing is documented.

https://github.com/artursapek/artur.co

Reply on news.ycombinator.com