Disappearing videos and disappointed grandmothers

https://rachelbythebay.com/w/2018/10/05/recipes/

by pjungwir • 8 years ago

251 104 8 years ago

irrational2• 8 years ago

This is why I don't trust my own videos/photos/documents/notes/etc. to "the cloud". How many examples like this have there been where peoples stuff just disappears one day with no way to bring it back? More than I can count.

Though, saying I don't trust my files to the cloud is not the same as saying I don't use the cloud. I keep two copies of all of my files in 2 physical location. However, both of those physical location are within 10 miles of each other, so I also mirror everything to the cloud. That is, I always have 3 copies of all of my files, 2 in physical location and 1 in the cloud.

I know so many people that have all of their important videos and photos in one location (their phone, the cloud, a hard drive in a drawer, etc.) I can't imagine how they sleep at night knowing everything could go poof at any time. I've tried explaining this to some of them, but they think I'm paranoid.

overcast• 8 years ago

Exactly, the cloud should be used as a secondary backup, not a primary device. Using it as the latter is no better than you keeping everything in one place at your house.

TeMPOraL• 8 years ago

Using it as the latter is no better than you keeping everything in one place at your house.

Not their house. Their neighbour's house. How long ago did this xkcd came out? https://xkcd.com/1150/

tzs• 8 years ago

I need to add cloud mirroring. What cloud service are you using?

I was about to give Backblaze Personal a try, but stopped because I don't like how they handle encryption. If I'm understanding their site correctly, restoring requires giving them your key. I'd prefer something where decrypting for restoration is client side.

I'm currently contemplating Backblaze B2 or Amazon Glacier using Arq or Duplicati[1] or something like that. (Or maybe using something I write myself which can better adapt to my specific needs).

Also eyeing OneDrive, since I have 1 TB of that from my Office 365 subscription, and am only using a whopping 300 MB of it.

Hmmm...could I make a large disk image and store the .dmg file on OneDrive, and then mount that image and use it as a Time Machine volume, thereby getting a cloud Time Machine backup?

[1] or duplicity or duplicacy...I'm not sure which of the three similarly named backup programs I'm thinking of!

Edit: looks like the answer to that last question is no. Time Machine does not seem to allow using a mounted image as a Time Machine volume.

uptown2• 8 years ago

For offsite backups I use Arq with Amazon, but the remote storage provider largely doesn’t matter all that much. Data is encrypted. Choose based on price.

kuzimoto2• 8 years ago

By the way, with backblaze you can specify your own key. They just can't help you if you forget, and you loose all your data :)

voltagex_• 8 years ago

Arq goes to OneDrive too. Just remember to exclude the Arq files from files that are downloaded by default if you're running the client. I'd look at B2 as well, but for the few terabytes I'd use it for it's a bit expensive with the plummeting Australian Dollar.

ahofmann• 8 years ago

I'm pretty happy with duplicati and backblaze b2 since crashplan stopped business with consumers.

irrational2• 8 years ago

I'm still on crashplan. I prepaid for 5 years, so I've been moved to the small business plan since my 5 years aren't up yet. At some point I'll probably need to look at moving my data somewhere else. Anyone have any recommendations on the cheapest place to backup 3TB of data?

northwest65• 8 years ago

Trust the cloud? Hell I won't even purchase music that isn't on CD or LP. If my access disappears for whatever reason (hacked, I've posted something politically distasteful on a linked service, whatever), even if I get it back, I could be a long time without my collection.

I have however accepted that as a possibility (no matter how unlikely) with Steam as I don't really go back to play old games, but I expect the music I purchase to accessible to me for decades without question.

vinceguidry• 8 years ago

People look at me like I'm crazy for not giving in to streaming services and insisting on maintaining a music library. It's hard for me to avoid concluding that these people simply don't really enjoy music.

cuddlecake• 8 years ago

That's sad

commandlinefan2• 8 years ago

someone who really gave a damn got their hands dirty

Although this person is the hero of the story, I feel obligated to point out that in today's JIRA-driven "what have you done for me lately" close-the-stories what's-your-velocity we're-about-to-downsize-so-you'd-better-prove-you're-valuable culture of "efficiency", the person who did this either had a LOT of job security (as in, compromising photos of the CEO in a public bathhouse job security) or did this on their own time while spending their "working hours" closing tickets and attending conference calls.

lkbm• 8 years ago

Or they work in a sane company that values individuals taking initiative?

Okay, maybe not a sane company, given the story, but there are plenty of places without the toxic culture you describe above.

commandlinefan2• 8 years ago

You mean a company whose management can tell the difference between working on something complex and not working at all? Sadly, very few exist, and those end up being bought by the evil kind.

ComputerGuru• 8 years ago

I’ve never worked at companies where using Jira was synonymous with bug quantity over quality. There will always be buglets, bugs, Bugs, and then BUGS.

On the flip side, every once in a while in the fish-shell GitHub repo we get an idiot that files an issue titled “is this even maintained?” because there is a non-zero number of opened issues (or I file a bug on GH and the the repo owner closes it immediately as “will revisit later” so the issue count doesn’t go up).

It’s not a technology problem. It’s a people problem.

userbinator• 8 years ago

https://en.wikipedia.org/wiki/Goodhart%27s_law

setquk• 8 years ago

It’s always a people problem.

You have to force people to own every ticket or they will just rot.

darioush• 8 years ago

#closing-tickets-and-attending-conference-calls #ctaacc (read like SEA-TAC)

I promise I'm sober

nkrisc3• 8 years ago

Remember: your users are people, not faceless lab rats.

domador• 8 years ago

Sounds like another outrageous case of digital sharecropping, where the value of the service was primarily provided by the users and the community they formed, yet such value was not respected, and people's work--a cherished part of their lives--was casually thrown away.

My more noble and magnanimous side would wish that unnamed service's owners and developers success in the face of such an awful mistake, while my harsher side feels like failure would serve them right for not respecting their users and not valuing things that should be valued.

itronitron• 8 years ago

Maybe we can file a patent application for a business process that discovers product features by learning behaviors within an online community. Even if we have to narrow the claims to product features that alienate the community I think we would still be able to find investors.

Full disclosure >> posting this here to establish prior art

mgkimsal2• 8 years ago

they didn't want to pay for storing all of those recipe and cooking videos twice...videos are big and space is expensive...

And yet... from people who never pay the bills for anything, all I ever hear is 'space is cheap, computing is cheap, blah blah blah'. It's all 'cheap' until you have to pay for it.

gonehome3• 8 years ago

In this case they destroyed the company instead of just saying 'space is cheap' and paying for it.

It isn't free, but it's certainly cheaper than killing your entire company.

saagarjha• 8 years ago

Well, they clearly didn’t realize that they would end up deleting their original data because they had tried to skimp on storage…

wstuartcl• 8 years ago

...

Not making a reference count and respecting it killed the company using one store of duplicate data, if anything, prolonged the company.

Nasrudith• 8 years ago

The people who say that are the ones who pay for programmer time which is much more expensive. Telling your programmers to erase all comments and noncompilation important whitespace would be monumentally dumb for instance. It is far cheaper to keep every version of code back to 1997 than to have to reverse compile and reverse engineer to document once for instance.

If you are getting anything out of it net greater than it and there isn't a better use for it keeping it is worth it.

mgkimsal2• 8 years ago

The people who say that are the ones who pay for programmer time

TYPICALLY the ones I hear that from are developers who aren't in charge of money/budget, and usually don't ever think about the cost of anything. The 'shiny/new' syndrome, coupled with little oversight, leads to nasty $ surprises later on.

And yep, "programmer time is more expensive" all night long. But at some point, other expenses are coming out of someone's budget someplace. Just because it's less than some other line item doesn't mean those costs can't add up to more than was expected (or needed, or budgeted for).

tetrazine2• 8 years ago

Well. Videos are orders of magnitude larger than many other common web content formats. And there are probably several codecs available on a CDN. It's reasonable to call that expensive and still claim that cloud computing is cheap overall. It is amazing to store gigs of photos for cents per month on AWS.

But I would guess bandwidth dwarfs storage costs here, so it does seem odd.

Cpoll• 8 years ago

But that's another good point, if you're storing multiple encodings, storing a single backup is less than double the cost, because you only need to back-up one encoding.

Slow-to-retrieve backup services like S3 Glacier are usually much cheaper, too.

theoh• 8 years ago

This is basic, though. If the stuff was stored in a Unix filesystem it would be the difference between a symlink (correct strategy) and a hard link (risky). Correctly implementing the hard link strategy (in the Unix analogy, this happens in the filesystem code itself) requires reference counting. Either you invest the time to implement that, or you should stick to symlinks.

farnsworth• 8 years ago

Anyone know what product she's referring to?

floatingatoll5• 8 years ago

Dear HN,

This post goes out of its way to avoid naming the site in question.

If you know the answer to the above question, think carefully before providing that name here or anywhere. Ask yourself how invoking their name specifically contributes value to this story. Ask yourself why you consider it right to shame them by name, when OP did not consider it right to do so.

If you can’t think of a reason that positively improves the world for us all, and all that you’ve got is “we should hold them up as an example for others to avoid”, this post already does so and without doxxing them.

If you truly feel that it’s necessary, proceed, but I think honoring the author’s clear intention to avoid naming them is the appropriate decision here as well.

mechnesium• 8 years ago

I respectfully disagree. It should be public record when a product or service fails to perform to the expectation of the customer. This is why we have outlets such as Yelp and Amazon reviews. There is no incentive to improve the product if they aren't put on blast.

tetrazine2• 8 years ago

Hmmm. Inclined to agree if you replace the company in this argument with a person.

Very disinclined in actuality, since this is a company providing a service to people who have handed over custody of their personal data (It sounds like very high-value data! Cooking videos aren't easy to make!) under the assumption that they will get something out of the deal, i.e., the videos would be hosted and they would be warned if they were going to be deleted (or at least apologized to if they were deleted by accident and not lied to).

>> If you can’t think of a reason that positively improves the world for us all, and all that you’ve got is “we should hold them up as an example for others to avoid”, this post already does so and without doxxing them.

Hmm, how about "we should hold them up as an example for users to avoid".

Doxxing doesn't exist for corporations, it only exists for individual people. The proper term of art here is whistleblowing or journalism. And the reason the author avoided mentioning the company name is probably to cover someone's ass who signed an NDA.

msla• 8 years ago

Ask yourself why you consider it right to shame them by name, when OP did not consider it right to do so.

To prevent people from getting ripped off.

Doxxing is the wrong metaphor here because it implies someone being put in a vulnerable position because of their private activity. It isn't "doxxing" to name the restaurant in town which persistently fails health department inspections, or the contractor who can't install shingles competently. Why? Because that's business, and business is in the public sphere. Providing bad reviews prevents those bad actors from harming other members of the public.

pkaye• 8 years ago

Seems relevent to know the company so we can avoid them in the future.

nkrisc3• 8 years ago

Please explain to me how a corporation can be "doxxed", because I don't understand how that's possible.

nraynaud• 8 years ago

I agree, you can't complain that every communication is locked and scripted and then try to hurt someone who has tried to be as candid as possible. It's just pushing towards a more and more unauthentic world.

jasode• 8 years ago

Ask yourself why you consider it right to shame them by name, when OP did not consider it right to do so. [...] ,

The other possibility that you didn't seem to consider is that the author (Rachel Kroll) is withholding the website's name to protect herself and not the company. If she only heard about the company's video deletion via second hand information, she'd understandably be hesitant to name the company in her blog. The person telling her the story may have not gotten the facts exactly right.

and without doxxing them.

If a user can factually state that <RealName> cooking website deleted videos, that's not "doxxing" the company. That's a very peculiar word to use and it doesn't apply to this situation. If someone factually states that the MS Windows 10 upgrade deletes users' files, that's not "doxxing Microsoft".

crunchlibrarian• 8 years ago

You can't "dox" a company by calling them out for lying and incompetence. Rachel simply does not want to get sued. This isn't some virtuous moral high ground you're standing on, stop pretending otherwise.

sbuttgereit• 8 years ago

I agree with you in general, but the other problem with the post that you're responding to is that it's likely to elicit guesses... and guesses in the guise of knowledge... some of which may be wrong.

Discussion like this that try to "name names" can result in people making assumptions about others (be they people or companies) that could very well be unjustified.

jaimex2• 8 years ago

Oh thats good! the Chinese government should use this one.

xaranke• 8 years ago

Cookpad?

ObsoleteNerd• 8 years ago

While I vehemently disagree with what the a user above you said about "doxxing" companies being wrong (it's not doxxing, it's whistleblowing), I also don't think we should guess. If someone KNOWS what company it is, I do believe it's in the interest of the uses and society to know this, as a warning, but guessing just potentially hurts innocent companies that have nothing to do with this.

castlecrasher2• 8 years ago

Even inside the company, the internal tech support is essentially gaslighting the employees who report this problem, by saying things like "it never existed"

It's mind-boggling that this happens but having experienced it myself, it sucks. Attempting to address an issue that sure seems to be due to another team's system is already difficult, so when they respond "it's not us" you're not only left holding the bag but questioning yourself for doubting that system.

perl4ever• 8 years ago

Some companies are caught in a cycle where they're not charging enough to their customers, much less enough for tech support, and as a result their policy seems to be to frustrate you as much as possible by denying everything if you contact them with problems, while refunding you store credit if you persist. Please send photos - you send photos - "we don't see anything wrong." A refund doesn't actually solve the issue!

Obviously, sympathy is limited for a free service, but I've encountered this with a paid service. And a few weeks later they announced broad based price increases, which made for a clearly implied backstory. Not enough money for labor costs.

ex3xu3• 8 years ago

Another even more egregious example -- around 10 or so years ago, I was a big proponent of the writing website Urbis. It was a site where you could submit a piece of writing and get feedback from other users. You can still see its activity on the Wayback Machine. It was an interesting site still without equal as far as I can tell.

From what I understand, one day, the creator decided that the site wasn't making him any money and shut it all down. I foolishly didn't back anything up and all my prose was gone. Granted, my teenage writing wasn't about to win any awards or woo any publishers. But some of the stuff I wrote was meaningful reflections on experiences at the time, meticulously edited and reviewed, and I still wish I would have been able to read them now.

What really irks me is that all he had to do was give me one day and I would have been able to save all my stuff. I have very little sympathy for any company that does anything like what Rachel describes here.

corey_moncure• 8 years ago

Did you at least get your money back?

leetcrew• 8 years ago

i'm assuming it was an ad supported service.

ex3xu3• 8 years ago

This was 2008, only a few years after Facebook had come out and well before the "People are the product" conversation that would come to dominate the discourse around monetization. Free services like Gmail and Livejournal had set a precedent for my naivete in expecting a decent product to be available for free. I've certainly learned my lesson from that experience.

bdr• 8 years ago

A little OT, but I’m curious to know what made Urbis better than the other options out there.

ex3xu3• 8 years ago

Probably the most unique thing about Urbis was their system for generating reviews which I thought was pretty interesting -- reviews on your work were locked by default. In order to unlock a review, you had to write a review of equal word count on someone else's work. I thought this was pretty successful method for incentivizing active participation on the site, something I haven't seen replicated anywhere else. Although I more or less grew out of writing recreationally, so I wouldn't even know if there were a successful modern alternative to Urbis.

JoblessWonder• 8 years ago

This reminds me of the issue that came up recently with Twitter (I think?) re-using image IDs that were very old. A lot of professional sports teams ended up with very odd/cryptic posts.

Here is an article with NSFW examples: https://deadspin.com/this-is-probably-how-the-cubs-ended-up-...

emayljames• 8 years ago

Thanks, I laughed very much at that article.

brennebeck2• 8 years ago

Wow. I was not expecting this. Ha, thanks!

aaaaaaaaaab• 8 years ago

Daily reminder that there’s no “cloud”, just other people’s computers.

gammateam• 8 years ago

oof gaslighting

anigbrowl• 8 years ago

Add to this youTube's annoying habit of leaving placeholders for deleted videos in my playlists but not telling me what it is that got removed, thus preventing me from ever finding it again. I fail to see what purpose is served by deleting the metadata as well as the possibly-copyright-infringing content.

chocolatkey• 8 years ago

Yes! If it at least said "xyz was removed due to a copyright claim by abc", but no, instead there are just blank spaces in the playlist. I leave them there on hope of this feature someday being added.

pmiller2• 8 years ago

Maybe the metadata is deleted on purpose to prevent you from finding the deleted video again?

abtinf• 8 years ago

YouTube is inconsistent about how it reports the video name. IIRC I noticed that on Roku, it will show the name in a playlist or if you try to play the video.

pandapower2• 8 years ago

If nothing else deleting/hiding the metadata makes it less likely that someone outside of youtube will figure out a way to accurately determine how many times a given piece of content has been uploaded to youtube. Instead they hide it all and pretend like it never happened.

thsowers• 8 years ago

I discovered this recently: If the uploader themselves deletes the video then it will also be removed from your youtube history, as well as your Google account's history @ myactivity.google.com

dabernathy89• 8 years ago

You can always almost find out by just Googling the video ID, especially if it's been submitted to Reddit or a similar link sharing site.

baud147258• 8 years ago

On my playlist with music that's not yet in the public domain, I've added in the description of the playlist the full list of the name of the songs. So that each time a video got removed, I only have to check the list and find another copy on YT. Still it only works because I rarely add any new videos to the playlist.

garganzol• 8 years ago

I spotted Facebook does this on more than several occasions. "Videos are big and space is expensive and we are cheap". Not only videos, but images would also spontaneously go for some users.

chungy• 8 years ago

Can someone explain what this "moments" concept is? I'm unfamiliar.

lifeisstillgood• 8 years ago

Surely there (should be) a duty of care over other people's data - and the loss here is actionable and costly.

I am a great believer in regulation through being sued to smithereens

kbenson• 8 years ago

This is reminiscent of the Reply All podcast's episode about a photo storage service that had major problems. If i recall correctly, some photos in that story were actually recovered, but it took a while.

...Okay, I did a little looking, and the followup to this episode was about a month later[2], and actually some pictures were able to be recovered after another company bought out the first and tried to save the photos. In any case, it's an interesting case study of an incident that shares someof the same themes.

1: https://www.gimletmedia.com/reply-all/71-the-picture-taker#e...

2: https://www.gimletmedia.com/reply-all/77-the-grand-tapestry-...

noonespecial• 8 years ago

Just another reminder that if its not on your disk drives(1), you don't have it.

(1) Yes, that is plural.

ap46• 8 years ago

I make it a point to Dropbox critical files(non-media) & do monthly backups of my phone & Mac to 2 physical drives. I'd love for something which does this daily of specific folders on my phone & Mac & makes sure there are no duplicates etc. And then on a monthly basis archive them to 2 or more different cloud storage vendors.

SZJX• 8 years ago

Not sure if the "moments" stuff is that "14-years-old". I'm not sure about the Snapchat implementation but the things on Instagram and FB etc. surely have people of all ages using them. Though yeah that's not the main point of the article I know.

Disappearing videos and disappointed grandmothers

Related Stories