SmugMug emails subscribers with request to find more paying users | Hacker News Reader

SmugMug emails subscribers with request to find more paying users

techcrunch.com

66 pointshellisothers58 comments

I have so many mixed feelings about this. I've always been a big Flickr fan. I absolutely loved the Explore section.

It had a set of really nice photos that I would spend every evening after working, browsing, sipping on a cup of tea and listening to music. I find some fun groups, some people who click some great photos on some really impressive gear which I don't own, but I would love to. The quality of Explore dropped drastically, sometimes showing up weird, creepy things, or something just random photos. It didn’t seem curated at all.

Fast-forward to recent times, I've seen Flickr bought and sold, schemes changed and more. There have been times of instability on the site, the Yahoo! login has been a pain. Performance has been erratic as a whole.

At a time where Instagram, Google Photos (and other cunning) services are the only healthy (in terms of users), ‘social’ photo hosting services, Flickr to my naive mind stands out so I don't want to see it go away. I started paying for a Flickr Pro to support it, even backed up my entire photo base to it. Now, I’m not sure if I need find alternatives

I feel the e-mail sounds more personal, than the usual two-faced PR spiel we see from larger corporations. It's nice that Smugmug appears to have good intentions, but it hasn't treated Flickr too well either. Logins have been broken in the recent past, there have been several outages, terrible migration attempts that almost speak of incompetence. Of late, it says I can't login using Firefox. Once logged in (after setting a different User Agent), it works perfectly fine.

So I really don't know what to feel.

I would have stayed with Flickr post-acquisition and likely even upgraded to a Pro account out of a sense of support.

That all went out the window when SmugMug decided to force a mandatory binding arbitration clause on its users, just like Verizon wanted to do to users who remained with OATH/Yahoo. I clicked on every link and followed every process I could to make sure my logins were closed out on either side and walked away.

I expect bullshit like that from companies like Verizon but not from "a thriving, family-owned and -operated business that cares deeply about photographers." Both companies should know better but I'm making the (obviously incorrect) assumption that SmugMug values caring more about respecting its users.

> Of late, it says I can't login using Firefox. Once logged in (after setting a different User Agent), it works perfectly fine.

This is icing on the crappy cupcake.

Haha. I thought I was the only one swimming upstream against binding arbitration. Just last night, after reading the absurdly long Zynga ToS update, I contacted them to opt out of binding arbitration. During the whole thing, the devil on my shoulder was saying, “You’re an idiot. This doesn’t matter. Let it go.”

Fuck em. Both Yahoo! and SmugMug have beaten the dead horse that is Flickr more than enough. I've moved on (and self-host these days) and I'm not looking back even if there were some inkling that Flickr was better at image hosting than I am.

For me, Flickr became less and less usable with every UI update. Back when Sonic was charging beaucoup bucks for 3Mbit DSL in San Francisco this was a big usability problem. The social stuff was neglected (SNR on the groups became abysmal). For a while I had a "pro" account. At some point they marked my account adult only with no explanation so I made another (free accout) and moved on. Then Yahoo! migrated everyone to Yahoo! logins and my Flickr accounts kind of operated in some odd grey area and I just use Flickr for free photo storage.

And then came SmugMug. My last public Flickr post was a screenshot of what SmugMug did to the Flickr mobile app — they'd just pushed some notification about the SmugMug buyout and all of the "excitement" clobbered the navigation links[1].

I tried in vain to migrate my Flickr account. The mobile app worked but wouldn't let me reset my password. The desktop site would send reset messages to my email address, but the auth infra thought I shouldn't be allowed to reset my password from that email address. Obviously the SmugMug guy who was posting on HN never got back to me. Neither dud Flickr/SmugMug support.

So I went back to self-hosting. And it worked. I didn't even have to beg support for access to anything. And now SmugMug wants more money? Sorry guys, you've had many chances.

1: https://live.staticflickr.com/979/41376824904_bba1b08851_o.p...

May I ask what software you use for self-hosting?

May I ask what software you use for self-hosting?

It's just a home grown thing. I'm not trying to compete with all the social crap that Flickr used to do well. These days I'm leaning towards making it more of a static site generator type thing. Image processing is done with a combination of imagemagick and exiftool.

Initially it started out as an Elixir+Phoenix app with an eye towards setting up some image processing as a set of background services taking the form of Elixir apps. Elixir and Phoenix were pretty painful so it's since been rewritten in Rust+Rocket.

Only four months ago, the same CEO said here at HN:

>I hope you'll stick it out with us, because we're nearly done migrating one of the largest web services on the planet, which means we get to focus on building again instead of just copying. The future is bright.

https://news.ycombinator.com/item?id=20649186

I've been a Flickr user for 13 years, and I can't say I'm surprised it's back in a precarious state. The sense that it could close up or drastically change (for the worse) has always been one of its defining characteristics. This is only the most recent in a string of disappointments following commitments to reinvigoration by successive owners. I remember telling Marissa Mayer, in person, thanks for investing in Flickr under Yahoo when they invited users to their relaunch [https://www.flickr.com/photos/nrbelex/8766466511/in/album-72...]. Almost immediately Yahoo lost interest. They made some minor UI changes, but the service as a whole languished. They probably came to the same conclusion: there's virtually no realistic path to profitability.

At this point, Flickr is simply too lumbering to become a competitor to Instagram. That's probably for the best. If they'd tried to make Flickr an Instagram clone, they'd probably have lost what few remaining dedicated users exist. The platform emphasizes the photos, and the community, or at least parts of it if you know where to look, really is focused on high quality images.

It's a particular bummer because Flickr is one of those services that's supposed to prove that a service can be primarily subscription-supported rather than ad supported if the product is good enough. If Flickr can't pull it off, even with a seemingly the best possible owner, what hope do other online services have? Flickr can and has jacked up pro prices--which is fine, frankly--but as this email makes clear, it probably won't be enough.

What I don't understand here is why SmugMug bought Flickr in the first place. Smugmug is such a great example of how to build a profitable photo storing/sharing-ish business, that to buy a service that is bleeding money without a path to break even seems like such an odd choice.

If you take their message at face value, they bought flickr just because they thought they should save it.

If you can't believe that, feel free to come up with any conspiracy you want.

I'm not sure what this adds. "They thought they should have it"? That just begs the question...

Save it, not have it.

OMG my bad. Thanks hahah

It’s so nice to see internet discussions resolved amicably. ️

Why AWS? It'd doubtlessly be cheaper to throw Flickr onto dedicated servers (by a ton). It would also probably help if they freed the source code of Flickr, so others could contribute and make it lighter to run.

A site like Flickr is exactly what you want to keep as far away from AWS as possible.

Stackexchange is probably one of the only companies that's doing hardware efficiently.

https://nickcraver.com/blog/2016/02/17/stack-overflow-the-ar...

Everyone else is more than happy to give more money to cloud hosting companies.

Somewhere on codinghorror.com, Jeff Atwood has a really good post on this.

Are you talking about "The Cloud is Just Someone Else's Computer"?

If so, it's available at https://blog.codinghorror.com/the-cloud-is-just-someone-else...

Dropbox, ServiceNow? ServiceNow goes so far as to give customers non-shared bare metal.

Backblaze is doing storage cheaper with their own gear.

SmugMug seems to be hosted on AWS (based on doing a DNS lookup and seeing who owns that IP). I wouldn't be surprised if moving it all to AWS was the path of least resistance to an engineering team that understands AWS more than managing a dedicated hardware fleet.

The SmugMug team were the first people I remember seeing use S3 for more than a 'hello world' project.

This blog entry from 2006[1] is only 13 years old, but feels like it should be older. Back then, putting that much data for a publicly accessible site was considered pretty insane.

[1] https://don.blogs.smugmug.com/2006/08/12/amazon-s3-the-holy-...

Older how?

S3 was invented in 2006.

What was S3 used for if not for hosting website data?

> Older how? > S3 was invented in 2006.

Because looking back at that, it feels odd that it's only 13 years ago, and how much the industry has changed in that time.

Back then S3 was this weird thing that Amazon came out with. A file hosting service that didn't let you do anything except store or fetch files.

Sure, it's low cost... but I can't process them without downloading them again and paying for the data access (twice!). Why the hell would I want that?

SmugMug's use case was the first I saw that actually made sense.

Back then, if you weren't able to afford buying a server and putting it in a colo facility, you'd go rent a server on a monthly/yearly basis from the likes of ThePlanet, and host those files yourself on your server. If you needed redundancy, you'd have to figure out syncing between the servers yourself.

Now? Spin up an AWS instance or trigger a function, serve a link to the file in S3 or via a cloudfront distribution. Even if you need to process the data before serving it - stream it straight out of S3, through your application and to the user.

At the place I work we transfer files with one of our vendors via s3.

I agreed. You dont even need crazy SSDs for all the data. You can have spin disks which have gotten cheaper for TB of data for the older images and SSDs for the cached ones (most recently accessed) and voila.

For write once archiving, bluray is an option too. Although who knows how long the discs last.

You're forgetting that engineers are expensive, and dedicated servers means spending more engineering time on things. Almost everyone considers hosting their own hardware, and it turns out that for most companies it's just not economical.

I really doubt they switched to AWS because it's "cool".

Also, isn't CDN bandwidth and data transfer going to be the lion's share of their costs? That's not a problem you can just throw dedicated servers at.

I'm not forgetting that at all: I don't see how a dedicated server takes any more time than AWS. AWS adds extra things you have to do and manage.

Also, data transfer and bandwidth is far cheaper with dedicated and colos than AWS.

My team runs a number of dedicated servers and a few cloud servers. AWS/GCP/the cloud do indeed add more things you need to be concerned with, but there are a lot of things you don't need to be concerned with.

We've had to fix: crappy datacenter networking, BMC issues left in place by the datacenter, hard drives spewing SMART errors, USB devices being plugged into our servers, bad RAM. Our datacenter company _moved_ at one point and we had a few hours of downtime as a server was driven to the new site.

Each of these isn't too bad, but when hardware issues or any kind of flakiness happens in "the cloud", it's so much easier to just bin the server and boot a new one – GCP will even live-migrate your VM as necessary to do this. That makes it so much more flexible. You can achieve this on dedicated servers, but not without a well resourced infrastructure team, which we (and many other companies of our size) do not have.

We don't believe we're going to save money moving to the cloud, we think it's going to cost more in fact, but we also believe we'll speed up quite significantly as a result of the move.

Physical space, building and upgrading the servers, monitoring and replacing components of the servers and their vast storage arrays when they are broken, upgrading the hardware, making sure all the software is up to date with latest security patches, reaching either end of life of their version and having to perform major upgrades with potential compatibility problems, cable management, all this at remote locations...

Vs. managing an Amazon account from a desk, any desk really, you don’t see the appeal?

Imagine you're looking for a new place to live. You go to a realtor and he starts showing you McMansions and telling you how great they are.

You say: "But those are really expensive, require a ton of maintenance, and hard to sell later on"

Your agent looks at you and says: "Yes but do you really went to find land, get the permits, get it developed, deal with all those contractors and deal with the inevitable problems of a new constructions? A McMansion is obviously better."

The agent has apparently never heard of apartments, condos, semi-detached and so on (yet somehow holds a very strong opinion). This is a problem; you find a new agent.

I can manage a Hetzner from any desk in the world. Same for basically any other server. Even if that weren't true, it's really simple to do dedicated servers, even if you're not fielding through a company like Hetzner. See this, from the founder of StackOverflow:

https://blog.codinghorror.com/building-servers-for-fun-and-p...

Dealing with security patches isn't anything you wouldn't already be doing with AWS, and is in some cases easier on dedicated hardware.

Someone from Hetzner sales team is going to be giving them a call soon. At least I hope so.

What makes Flickr more suited for dedicated hosting rather than awe?

Bandwidth is expensive on aws.

Image and video hosting is usually significantly cheaper on bare metal where you just get unmetered bandwidth from housing providers

You don't know what bulk discounts large customers negotiate. High egress prices are a way to get a few extra bucks off small-potato hypermilers

You're right, I don't know their discounts.

It's still more expensive unless they get paid for using the bandwidth though

Money?

Where are former customer going? Where are people looking to host photos going nowadays? Just want to survey HN.

Personally, although I am relatively hardcore on privacy, I go with Google Photos on an anonymous account (I know that since it's Google, nothing is truly anonymous, but from the steps I've taken hopefully it's a bit more sanitized) because their AI can filter and organize my photos is ways that I simply don't have the time to do.

I got sick of site after site closing down along with my albums, or, worse, using my metadata to feed the panopticon, so I quit my job to build an alternative. It's early days still, but ML auto tagging is on the horizon. Runs on desktops, servers, and within docker. Details in my profile.

Join up with this person https://news.ycombinator.com/item?id=21842347

We don't need more half-attempts.

Thanks for your suggestion.

PhotoStructure already has the best metadata inference and asset de-duplicating system that I'm aware of.

It scales to millions of asset files, works cross-platform, handles JPG, RAW, and transcodes videos to web-browsable formats automatically.

I haven't used an open-source project yet that didn't fall over after the first several thousand imports, and none de-dupe (not just the same SHA, mind you, but the semantically equivalent image, based on both metadata and rotation-invariant image content). They also tend to be abandoned when the original author moves on due personal reasons, toxic comments, or the "v2 rewrite" effort collapses under it's own weight.

This looks promising, thanks for sharing. Enjoyed reading about how you use metadata to tag everything and make it easy to find related photos.

After more than a decade as a Flickr pro user I've canceled the account and downloaded all my photos. They're now hosted on a synology NAS which has decent photo management apps I can host myself without worrying about data loss and privacy. Then that NAS backs up to backblaze B2 for offsite redundancy.

I realised that I was spending a lot of non-productive time on the 'social' aspects of Flickr - essentially commenting on other people's pictures to get views on my own. This effort had essentially zero useful payback in the real world and is prime reason for not moving to (and paying for) SmugMug or other portfolio sites. My dislike of the panopticon / surveillance capitalism means that I haven't attempted to move to the other obvious social media platforms (Instagram, Pinterest, etc).

So, I am still taking photographs, but have no desire to share them with others and get into the inevitable competitive desire to get 'best picture'. My 600k views and several Flickr explores (several years ago) have cured me of that itch.

Personal Photo Management powered by Go and Google TensorFlow - https://demo.photoprism.org/

[] https://github.com/photoprism/photoprism

That demo works quite fast. Can you please let us know what kind of hardware are you using to deploy this? Any secret sauce to make it load fast?

It’d be very sad to see Flickr die. I started amateur photography many years ago and immediately signed up and even paid pro account for years. After I got busy with other stuff I stopped paying it, but still is my main place for photos and even sold a few just because of the exposure.

I guess it’s getting a bit desperate if they have to ask for subscriptions via email. Also the fact that they moved to AWS for a service that depends so much on data transfer makes me think it’ll get even tougher to survive. I hope I’m wrong.

I have received the email yesterday, been a Flickr user for 10 years. I had mixed feelings as I really like Flickr but at the same time how would the subscription save the company? The email didn't mention numbers of pro accounts required to break even or reduce the loss. At the same time, I thought we they didn't put some adverts for non-Pro memebers to test if this would help to provide a new income stream which might worth more than a Pro subscription.

https://www.archiveteam.org/index.php?title=Flickr

I renewed my Flickr Pro subscription for the first time in a few years after SmugMug's acquisition, but I barely ended up using the site, and let it lapse a month or so ago. It bums me out to say it, but I think photography might have moved on from Flickr. I know mine has, at least.

But, where? I’m genuinely curious about alternatives.

Insta is a like driven mess, with almost no possibility for constructive feedback, thoughtful discussions about photography, critique.

500px had a decent UX, incl. galleries, comments, but quickly became a formula-driven, Instagram-like monster.

Reddit... has decent communities, but it’s features are limited, the right UX just isn’t there.

Is this a niche that can be filled easily (community vs. black magic/algo-driven discovery, archiving and photo critique)? Or, is this just a really hard problem to solve in a sustainable (financially) way?

None of the above. I found a local darkroom and a group of likeminded photographers with whom I have regular critique sessions. I don’t find online critique to be useful for the directions in which I’ve been taking my work over the past couple years.