Hacker News new | past | comments | ask | show | jobs | submit login
Less than a month to go before Google breaks links to Google+ Picasa albums (greenspun.com)
100 points by scarhill 17 days ago | hide | past | web | favorite | 90 comments

In Google's defense:

1. This was announced over 3 years ago [0]

2. Photos/Videos uploaded to Picasa were transferred to Google Photos automatically (comments/tags/captions will be lost though) [1]

3. Picasa was originally built in 2002! (bought by Google in 2004) -- Perhaps Picasa had a time and place?

[0] https://en.wikipedia.org/wiki/Picasa#Discontinuation

[1] https://picasa.google.com/

Yeah, I'm often one to criticise Google, but an entirely free cloud-based photo album was never going to be around forever. If you are very concerned with keeping an online record for all time you really shouldn't be depending on free stuff like Picasa.

Maybe, but it would also cost Google almost nothing to set up a 301 redirect from Picasa URLs to Google Photos. I wouldn't expect someone like Yahoo or Microsoft to bother, but being a “good web citizen” seems intrinsic to the Google corporate identity, after all.

The harder it is to find things via following links, the more you’ll resort to search. Google has a perverse incentive to break the links.

But that doesn't solve the issue that the photos on Google Photos are private by default, so Google would need to change the sharing to public on all photos that have external links, which would be a terrible privacy measure.

> an entirely free cloud-based photo album

Uhh.. isn't that exactly what Google Photos is?

Hello Gmail users! :)

You really shouldn't count on having Gmail available forever!

Or even for a few years.

If you use Inbox, you can't count on it being available for more than few weeks :(

It's interesting that Amazon's approach of building products and killing them off if they don't improve the bottom line earns praise, while Google's similar one endlessly pisses people off.

It's probably because by and large Amazon kills physical products and Google kills services.

If someone buys a Fire Phone and it gets discontinued - well, products getting discontinued is part of life and it doesn't (usually) affect the usage of the one you have.

If Google builds a web service, then that service has the chance to grow to 50m / 100m users before deciding that it is not worth supporting. Law of large numbers says of those 50/100m a non-trivial proportion will invest substantial time in it and get pissed off when they have to make alternative plans.

If Amazon killed Prime Video or Audible, I bet you’d see a pretty huge outcry.

Amazon ended music uploads [0] But - yea, no HN outrage...

[0] https://www.amazon.com/gp/help/customer/display.html?nodeId=...

Wait until Amazon starts killing off some AWS services or features.

Which is likely why they have gone out of their way to not kill off AWS services.

I wonder about the lifetimes and user engagement between examples from Amazon and Google. It's interesting that my off-the-cuff impression is that Google gets into hot water here by letting its experiments 1) run for a long time, 2) acquire a lot of users because "hey, it's free" and 3) sometimes suck the air out of a product segment because #2 (see: Google Reader). Likewise, my impression of Amazon's similar experiments are that they're cut short at a much younger age, with lesser engagement. Now I'm very curious: does the data actually back up those impressions?

Maybe I'm wrong but I think the difference is Amazon always/mostly charges the primary users money. Googles business seems to be 'hey it's free' for primary customers, charge secondary customers $$$. Problem with that is much later when it becomes obvious there isn't enough money to be made by spying on the now quite large user base. Google then just tosses the thing over the side.

Mostly because people don't really use any of the products Amazon releases at scale. There's a reason why it's called Google scale and not Amazon scale.

Also because as humans, we're generally more inclined to dogpile onto the Goliath and root for the David. Underdogs almost always get treated more leniently.

>Mostly because people don't really use any of the products Amazon releases at scale.

AWS would beg to differ.

However, Amazon has been very good at not getting rid of AWS services so it's reputation is that it's at scale services are going to stay safe.

Maybe to clarify better, it's the bunches of ancillary products and services I'm talking about. Core items like AWS and Google Cloud are of course by definition going to require an inherent scalability to be meaningful.

AWS is not a single service but a large collection of separate services that interconnect. Some of which are quiet old and not exactly cutting edge. Yet they don't get deprecated.

You could say the exact same thing about Google Cloud. What I'm more talking about is stuff like Google Keep, Docs, Slides, Sheets, Kubernetes, Istio, etc. etc.

For what it's worth, Amazon is the same size as Google now (technically a little bit larger by market cap).

Google doesn't really have much of a retail division though, and is almost entirely services-based. When services disappear, the time and effort you've invested into them disappear.

When a shop stops selling a product, you can still keep using the product you have. They don't take it away from you.

I think it's because people like Google services, but never even heard of Amazon's services.

Amazon's are probably almost not used at all, whereas Google kills off products that still have plenty of users, even if not by the standards of Search or Gmail.

This is probably by design on Amazon’s part, in which case the lopsided outrage is entirely justified.

Like which? I'd never noticed them doing this, which may mean they were never that great in the first place.

> while Google's similar one endlessly pisses people off.

Maybe it's because we know Google is profiting off of our private information, and we don't make a stink about that because they're giving us free stuff. But why shouldn't we be upset when they stop giving us the free stuff we like? They're still profiting off of our private information aren't they?

Amazon just isn't in the same boat.

What if they simply don't profit enough to justify giving you free stuff?

I never said they were wrong to deprecate unprofitable services, I'm just saying that it's not like the outrage is coming from nowhere, or has no justification.

Wave could have been UUUGE, though.

I'm not sure. Wave has been out (and discontinued) for a LONG time. It was even transfered to Apache under an open source license. And it's never taken off. No one has ever picked it up. No one has seriously tried to replicate it. If there was a business there, if there was a use case there, someone (anyone) had plenty of opportunity to take it and run.

The fact that not only the idea but also the implementation and the hype and the marketing were all available completely for free and no one ever turned it into a real product tells me that it wasn't a real product.

Wave needs a critical mass of users to be useful so that created difficulty with no existing (public) service after Google shut their instance down.

Google completely botched the roll-out. Wave is essentially useless without others to use it with (no value above Google Docs to use it solo, or standard email/IM to chat with others). I had 10 invites, I sent most of them out. Only two people of the 7 or 8 I invited received their invites. It was completely useless to me because one of the two was someone who just wanted an invite (but I didn't interact with much), and the other was an fellow tabletop gamer and the rest of our group couldn't get in.

When Google finally opened it (looked up dates, I remembered it was short but didn't recall it being this short) to the general public it was May 2010. They announced that they were going to stop developing it (switch to sustainment with plans to shut it down) in August 2010. They gave the service 3 months, and said it didn't have sufficient interest.

No shit, it was three-damned-months!

As best I can tell, someone at Google had intended to shut it down from the start. The fact that their attempt to launch it failed doesn't mean that the system (or its concept) had no potential. But killing it did kill the potential, because you needed that critical mass to gain any value from it.

It had federation which would've been useful, but with no public instance there was little reason for a small group to run their own instance. There was no one to talk to.

I agree and disagree. The fact that a giant company tried it and failed, and the fact that Apache Wave just sits there unused both suggest that there likely isn't a business there, but there have been many occasions where a company markets more or less exactly what has been tried unsuccessfully before a tad differently and it suddenly takes off like wildfire.

I would not at all be surprised if something Wave-ish became a huge deal someday, not would I be surprised if a dozen companies tried it and failed before someone makes it work.

Wave was what Facebook is. No one jumped.

Really miss picasa. I know the world is moving away from tagging your photos with descriptive words, but I still do it.

As with any Picasa related subject: Does anyone have a viable alternative desktop software that allows tagging of photos, some mild editing and allowing photos to be kept in folders that are the albums.

Picasa's flow is still the best quick-edit system I've used. I am comfortable with Lightroom & Photoshop, and they're great tools. But Picasa did a stellar job of balancing simple/lightweight/fast/useful in a manner I haven't seen replicated anywhere else.

I'm glad somebody said this. I totally agree. I still use the Picasa desktop software. It's brilliant for picking out best pics and tweaking them. The web interface just doesn't compare.

When they announced that the desktop software was going away, I immediately downloaded the final version and archived it away. More than likely I will use it until Windows no longer supports it... unless someone "makes a better Picasa".

It probably has only gotten more responsive by virtue of being made to run fast on hardware from the early 2000s

You're not wrong. I'll be genuinely sad when Windows eventually drops whatever API support allows it to keep running without emulation.

Microsoft is famously unwilling to let software stop working. Picasa will probably work forever.

Dropping Win16 was a bad sign. https://www.joelonsoftware.com/2004/06/13/how-microsoft-lost... talks about the back compat faction losing power at Microsoft.

Interesting read. A lot has changed in the almost 15 years since he published it. Has he done any followups?

It's not free, but Lightroom is the canonical software for this purpose.

I used lightroom too. I'm on the last version that isn't subscription based. Since I didn't upgrade my camera, I could stick with the old version (Adobe doesn't back port new cameras into the old version of its software). I noticed the "view photos on map" feature is now broken.

and since my new desktop is linux I've been looking for alternatives. "DarkTable" seems promising but I haven't really got the hand of the UI.

Darktable is a bit laggy (sadly), but it has a whole bunch of amazing filters that not even the latest Photoshop or Lightroom has, so it is well worth diving into. The equalizer tool is incredibly powerful and flexible, for example[0].

That linked video is by Robert Hutton, he makes really good tutorials for Darktable so that might help you.

Anyway, for non-realistic stuff you can add G'MIC for more filter-options and you are pretty much set for life[1].

[0] https://www.youtube.com/watch?v=zzVXK4eAM5E

[1] http://gmic.eu/

I want to use and like Darktable, but it just feels so foreign after using LR for over a decade at this point. I keep giving it a try, so maybe one day it will stick :)

Oh believe me, I totally get the feeling - muscle memory can backfire like that. Especially if neither old nor new interface is perfectly intuitive, and also just similar enough to be confusing where they differ.

I can't stress enough how good Robert Hutton's tutorial videos are though! They often contain little nuggets of behavior that I missed when first learning a new tool. For example, until I watched his videos I didn't know double-clicking on any curve resets it. Knowing those kind of little things adds up over time.


Why isn’t there a move TO tagging photos with text? That rocks!

This is all so depressing. Picasa desktop was great. They killed it. Picasa Web albums were great. They killed it.

Why are they making everything worse?

Because the number of not-willing-to-pay, enthusiast, micro-managing, photo-organizing users is likely tiny, and not enough to support a product with Google costs.

The market is splitting more and more into consumers who take too many photos to organize it themselves and pros who need robust workflow management solutions.

Don't get me wrong, I loved Picasa, and was very sad when they stopped supporting.

I particularly miss some features, such as creating a time-lapse of all photos of a certain person with the face aligned, which was awesome to show how friends and kids change.

To be fair, I've had surprisingly good luck at searching for photos in Google Photos by text. The AI there seems pretty solid.

I've had the opposite luck. Even typing something as simple as dog misses a ton of photos with a dog in them. Perhaps I missed a switch in configuration along the way to allow aggressive AI recognition. :)

It's very inconsistent. Sometimes I search for dogs and get dog fursuiters, but not actual dogs. It does pull up fursuits if I search for them though. Shout out to the furries working on the Google Photos team.

Flickr, the progenetor of photo tagging, remains a great place to do that.

Desktop in 2019? good luck. Adobe or Apple.

digikam is the closest I have found. UI is not nearly as clean and simple.

Google may not be much worse than other FANGs, but the way it pretends to be friendly to open source and free culture communities and co-opts them is nasty.

"Despite illusions in 2008–2009 that it was a fair player, Google is now trashing free culture by making all the Picasa Web images in Creative Commons vanish from the web.

In fact, users and albums are often forced to "migrate" to Google Plus, without telling them that any Creative Commons marking will be irreversibly destroyed in the process. There's no way to mark Creative Commons images on Google Plus. There's also no way to search or browse Picasa Web by license, apparently (the feature used to exist in 2009)."


If you care about free culture, don't put it up on a private (company) service. Most things on the web have not guarantee of longevity. The only thing open source about Google is some of the software licenses they use.

> pretends to be friendly to open source

I quite like go and Kubernetes.

That was meant to be parsed out as "open source communities".

When it's in their own self-interest I'm not sure whether they're being friendly or not. Can't really tell.

Also go feels a bit tightly controlled to me and like it's not carrying on the traditions of open source scripting languages (same with Swift). Fortunately, Rust and Elixir are shaping up to be good continuations of the culture of Perl/Python/Ruby/etc.

> When it's in their own self-interest I'm not sure whether they're being friendly or not. Can't really tell.

Does it have to be mutually exclusive. Sure, Google is hoping you're going to go GKE for your Kubernetes/Container needs. But you're completely free to use another cloud, or roll your own.

For all cloud native backend stuff I really love Go (and Rust for that matter).

I wouldn't depend on Google for anything.

Maybe self-hosting needs to be made easier for the average person? The web has so much bandwidth now that there's no reason outside of potential security issues that people can't be posting more of their own stuff(like a web).

It's far easier, more scalable, and more secure now than it's ever been, and people quite simply don't want it. I wish it weren't so, but people prefer staying inside walled gardens to homesteading.

They do, but I think there's some potential demand for it.

For instance, I want to be able to host my own files without having to make some sort of deal with my ISP. Although I've never had a problem serving requests directly from my computer, it's usually against the rules of ISPs. With a regular internet connection, I should be able to more easily store files on my own hardware and be able to share them with people as easily as I could with Instagram. Maybe there could even be an always-on hardware device that integrates in an open way with third parties that can serve those files, but ultimately those files live on that hardware and can be designated their own domains for URL resolution. Peer-to-peer and Cloudflare support could come with it out-of-the-box to mitigate load issues.

Meh, maybe nobody wants anything like that. I still like the idea.

I want something like that, but if I'm asked to shell out $50 for a box that does what I can do on Imgur/Instagram/Photos for free, I'm going to turn my nose on it.

With 99% of the images I want to share on the web - I don't really care if they aren't going to be around in five months time.

Also, what if the vendor of the box (That handles the integrations with third-party hosts) goes out of business? Am I going to now have to figure out how to hack my hardware to be allowed to talk to <Whatever the new image hosting service> is?

This is an inescapable rabbit hole of third-party dependencies.

i have the additional requirement that it survives my death ... which rules out anything one has to pay for, or ping every x months, and self hosting (even ones house disappears on ones death).

The solution i have found so far is GitHub Pages. It seems the website, though static, can be arbitrarily deep and large.

Self-hosting has become shamefully easy now with Docker Hub and cheap VPSen in the "cloud".

It really is a one-liner to spin up a new service on your server nowadays.

I mean, it is a one-liner, but only if you know what to do with that line -- linking to a URL in a message or putting a URL in the URL bar is the most complex one-liner you can reasonably require non-tech types to do, in my experience.

Technology that aims to serve that audience should be reasonably usable for that audience.

Point me at the self-hosted feature-equivalent Picasa clone that my grandmother can use- she still mourns the demise of Picasa- and I will set one up for her.

Me too, please!

If you know what a console is.

And you have a server.

And have a domain name

And backups

And time to apply security updates frequently and promptly.

and a credit card to pay monthly fees (which may or may not increase dramatically)

tell that to my mom

Add this to the list of reasons why you should not rely on cloud providers for anything you actually care about.

Maybe the right way to use Cloud providers is to develop against a "lowest common denominator" of functionality among the big kahunas. Then when of of them breaks, moving your stuff over to one of the others isn't that big a deal.

Be very skeptical of any "differentiating" feature that a Cloud provider tries to sell you.

There's no obvious minimal subset of features. Here's a feature Google Photos doesn't have: you cannot automatically sort photos in an album based on the filename. Either you sort by hand, or you sort based on the EXIF data. Picasa could, but Google Photos never got that particular tiny feature.

It means if you organize photos with a program, export them so they sort correctly based on filename, you'll lose your sort as soon as Google Photos gets its ugly hands on them. Google doesn't understand that EXIF times can be wrong.

On a slightly related note, what do people use to save local copies of valuable Google+ groups? If I understand correctly we're less than a month from shutdown?

Maybe someone could maybe a domain like fixedgoogleplus.com (similar to rawgithub.com) and host people's Picasa/Google+ photos while keeping the same link

When Google will be shutting down Google Photos?

Of course, it is going to happen eventually.

However, I do have a dozen of photos there with no easy way to export hiqh quality photos in one go that worries me.

Exporting is easy. On desktop you can select photos and download a zip. You can also download them with Google Backup and Sync if you enable the option to show photos in Google Drive.

There is a limit thought when I tried to export from web some time ago. It was 500-2000 photos or something.

I probably have over 30000 photos there so it's a long process to download them all.

The desktop sync client is probably the way to go then.

Google takeout was great for this. I did it last month and backed up 60gb of photos.

Picasa was such a great project. Google ruined it.

As George Lopez would say, Why you crying? You chose to use someone's free platform to host and display your pictures with virtually no work on your part, and now you don't get the free lunch anymore.

These scenarios are actually useful. They train people to use solutions that are open and portable, which will help them respond to unexpected disasters in the future. Use a self-hosted CMS to build, manage and publish your content so you can re-publish your content somewhere else.

> These scenarios are actually useful. They train people to use solutions that are open and portable, which will help them respond to unexpected disasters in the future. Use a self-hosted CMS to build, manage and publish your content so you can re-publish your content somewhere else.

If you're able to point to an open and portable self-hosted CMS that has an equivalent feature set to Picasa (and not a minor subset thereof), you'll be sure to gain a lot of praise from myself, and many others here too I'm sure.

I don't know why you got modded down, but this is the KEY point here. Yes, I loved Picasa and am pissed when Google drop cool services, but overall it is a good thing to make everyone more aware that their data is not permanently safe "in the cloud"

Applications are open for YC Summer 2019

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact