[HN Gopher] Show HN: Myphotos.site - Make a gallery website from...
___________________________________________________________________
Show HN: Myphotos.site - Make a gallery website from Google Photos
Hey HN! My friend and I love taking pictures on our phones, and we
were looking for a super simple way to share them with the world.
Google Photos albums are nice, but the result isn't slick enough
and lacks customization options (embed, custom domain). That's why
we built myphotos.site - a micro SaaS that allows anyone to create
a slick, minimal gallery website from Google Photos. Galleries can
be iframe-embedded for those looking for full customization. One
thing we haven't solved yet - many Google Photos users (myself
included) do not have images pre-organized in albums, so the fact
that _technically_ it takes 60 seconds to build a website doesn 't
take into account that you have to spend time picking the right
images. Would love to some feedback/ideas from hobbyist
photographers here!
Author : gabigrin
Score : 135 points
Date : 2024-11-28 11:02 UTC (11 hours ago)
(HTM) web link (www.myphotos.site)
(TXT) w3m dump (www.myphotos.site)
| mtmail wrote:
| Great website and the pricing makes sense. Some links still go to
| gphotos.site (automatic redirected). I tried so see how easy the
| iframe feature is, or rather if a non-technical person would
| understand the documentation. But the guides seem to be mostly AI
| generated SEO. Quickly the good impression of personal photos
| from genuine authors of the website turned into gen-AI slob
| https://www.myphotos.site/platforms/wordpress
| gabigrin wrote:
| Thanks! Re: links, will fix it. That was the original domain
| but Google didn't approve it, and we can totally understand
| why.
|
| Embed - it is a feature we pushed yesterday so still no proper
| documentation on that. It's as simple as importing images ->
| clicking on an "embed" button and grabbing the iframe code.
| Requires some technical understanding. We might improve to be
| platform-specific (Webflow, Wix) if there's a real need.
|
| And as for the guides you're seeing are complete gen-AI slob, I
| agree. Not something we're that proud of, but mostly for SEO to
| be able to acquire users at a reasonable price. We're running
| ads campaigns but with our low pricing, it'll be really hard to
| be profitable so SEO is one way.
| have_faith wrote:
| I work on landing pages occasionally that have to account for
| SEO so can definitely sympathise with the uphill battle here.
| Like the OP, I was immediately put off of the "brand" when I
| saw the flood of AI stuff. It's like seeing how the sausage
| is made.
|
| I think there's an extra level of trust that needs building
| from a branding perspective because of how personal it feels
| to hand over personal images to someone to host for you, even
| if they become public anyway. It's interesting to see how
| people react to this sort of thing as it's very new.
|
| Good luck!
| orlevco wrote:
| Co-founder of MyPhotos here,
|
| You're right about trust being important. We're working
| hard to show that we're real people with good intentions.
|
| I actually didn't consider that using AI to generate some
| blogposts (that aren't the main content of the site) would
| have negative impact on how people perceive the product. So
| thanks, we appreciate the feedback and will take it into
| account moving forward.
| wrycoder wrote:
| The problem for me is not the SEO, but the vast AI
| blather, which is virtually content-free. You need a
| complete rewrite by an experienced human. After reading
| and reading, all I found was subjective comments about
| how great all the features are, without any information
| about the features themselves. What is your relation to
| Google? Can photos only come into your site from Google?
| aziaziazi wrote:
| > I actually didn't consider that using AI to generate
| some blogposts (that aren't the main content of the site)
| would have negative impact on how people perceive the
| product.
|
| As a cofounder you already know your product. I don't,
| and one quick and easy way to discover is crawling your
| site (testing your product comes second as I need a bit
| of confidence before sharing my google credentials). Bad
| copyrighting (ai gen or not) depict one interest of
| having more customer while not much interest in sharing
| useful information on the subject. I don't know about
| your motives so I can only trust you "are real people
| with good intention" and wish you good luck, but as an
| anonymous visitor I surely would have passed my way.
|
| If the guide posts aren't mean for humans but the robots,
| maybe hide the link in a SEO friendly way? 10px off
| screen or z-indexed bellow another block for exemple.
|
| Also a header disclaimer would be very appreciated:
| "Content for SEO, back home"
| sigseg1v wrote:
| I think it depends on the target audience. For casual
| photographers many might not care about AI but poor
| curation makes it look cheap. For artists, if they see
| any hint of even minimal AI usage or hints that you may
| approve of gen AI it would be a total dealbreaker. For
| the subset of tech minded people that approve of AI, they
| probably already know how to host a site -> but maybe
| they are still a viable audience if this saves them a lot
| of {time|money|maintenance}.
| salomonk_mur wrote:
| SEO has changed a lot over the past year. It is now very hard
| to be featured on the top of the first page even with the
| best content. Several colleagues have seen their search
| engine traffic plummet 50-80%.
|
| I recommend going for other marketing strategies such as
| collaborations with bloggers and influencers in your niche
| (photo enthusiasts?). Better targeted, less vulnerable to the
| whims of Google's algorithm.
| swyx wrote:
| > And as for the guides you're seeing are complete gen-AI
| slob, I agree. Not something we're that proud of, but mostly
| for SEO to be able to acquire users at a reasonable price.
|
| no, this is a skill issue. follow fofr_ai to get tips on
| getting realistic photos. its not hard, you just havent
| tried.
| gabigrin wrote:
| Yeah, you're right. We're gonna fix the images ASAP. They
| are indeed embarrassing. Also, the term "guides" is
| misleading as it sounds like documentation, which it isn't.
|
| Thanks for the feedback!
| dinosaurbone wrote:
| I would be concerned that it also flies a bit too close to the
| sun with the gen AI photos; if one of the primary markets is
| creatives who are trying to showcase their (non-AI) photos, it
| doesn't inspire confidence to see the company they're
| entrusting with that task to seemingly disregard the field
| they're in.
| cryptozeus wrote:
| Agree ! Please remove those tacky ai stock photos !
| darubramha wrote:
| Looks awesome, and will help quick galleries to be created.
|
| I wanted to build something similar, but at that time photos did
| not expose APIs to get list of photos.
|
| What was your work around? Or have they opened up the APIs now?
| Alifatisk wrote:
| Curious about this aswell
| mxuribe wrote:
| I'm not affiliated with this effort, but would the following
| google api suffice? https://developers.google.com/photos
| gabigrin wrote:
| Thanks! We're using the Google Picker API -
| https://developers.google.com/photos/picker/reference/rest You
| create a picker "session" which is a link where the user can
| select images from Google's side. Then you poll the session and
| once the user is done, you can get access to the media items
| (imgs and videos) that the user selected.
|
| At first, we wanted to auto-sync galleries and were bummed by
| the way this API works, but tbh it ended up being more private
| and secure for the user (no auto-syncing that embarrassing
| picture that shouldn't be online) and helped us avoid
| implementing a picker on our end.
| gnaman wrote:
| How long is the session active for? Do you store these photos
| on your servers?
| gabigrin wrote:
| The session is active for couple of hours IIRC. Once the
| user is done with the selection we upload the images to our
| servers (Supabase storage) and delete the GPhotos session.
| sigmar wrote:
| probably entirely google's fault, but I can't seem to pick a
| folder when I'm brought to the google photos search box. it only
| finds individual photos
| gabigrin wrote:
| It's far from ideal but if you search for an album, you can
| quickly select all of it. Here's a video of Or, my co-founder
| showing that -
| https://youtu.be/7gi58SuQ6Rk?si=LQu8fJBDflVR11Rs&t=38 (links to
| the exact second)
| croisillon wrote:
| as someone who has been involved in gallery making in the past, i
| love it
| orlevco wrote:
| That's cool to hear! Was there anything specific that sucked
| about how you had to do it then?
|
| We'd love to hear more to understand what problems we could
| solve better.
| croisillon wrote:
| i think my main problem was the balance between
| overengineering (like the possibility to have white borders
| and shit) and making it user-friendly
| gabigrin wrote:
| sounds relatable :D
| ddtaylor wrote:
| I appreciate the presentation and simplicity. You did a good job
| of presenting it.
| yapyap wrote:
| the website itself looks very Template-y, now if that's not of
| concern just ignore I said anything but that's always what comes
| to mind when I visit a new website, like anecdotally I also hate
| it when a website has gone the whole extra mile to customize
| their website and then you check out and see the monotonous,
| bland things that are the stripe checkout buttons & page. Don't
| get me wrong, I get why from a backend perspective, totally. It's
| just takes you out of the whole immersion in the website (design)
| gabigrin wrote:
| The "give users full control" vs. "help the user reach a fine
| result" tension is real. Our take on that is allowing pro-users
| who want full control to embed their gallery in their website.
| We can't nor want to compete with full-blown website builders
| on customization features.
| cj wrote:
| Awesome nice work!
|
| Does this work for video?
|
| I recently digitized about 80 home videos from 30-40 years ago.
| Each of the 80 videos is 2 hours of 5-10 minute recordings from
| the olden days when turning "record" on and off didn't
| automatically create a new video file. I've been wondering the
| best way to post-edit and share them with other people in my
| family.
| orlevco wrote:
| Co-founder of MyPhotos here,
|
| First of all - thanks!
|
| Re: Videos - we currently don't support them but might add that
| capability in the future. Videos complicate things both in
| terms of the product and because they're much more expensive to
| store, so they'll complicate the unit economics.
|
| Regardless - online video storage is currently very expensive.
| The leading platforms are Vimeo and Wistia, so you might
| consider them for now.
|
| About editing - what are you looking to do? I'll try to point
| you in the right direction.
|
| Or
| atonse wrote:
| Check out Cloudflare Stream and see if their prices might
| work.
| gabigrin wrote:
| Will do! thank you
| indigodaddy wrote:
| You could get uncloudy and go with a provider with unmetered
| bandwidth and then use their storage slabs which present as
| block devices up to 10TB. So you could string together a
| PV/VG with eight 10TB slabs on a single server.
|
| I'd suspect the storage cost of this approach might even be
| similar or less than AWS glacier storage..
|
| BuyVM has been around for a long time and very stable and
| excellent reputation. You'll find that your costs would be
| much lower if you'd "uncloud" (I'm obviously assuming that
| you are using AWS or a cloud, but maybe not!)
|
| https://buyvm.net/block-storage-slabs/
| mceachen wrote:
| Thanks to ChatGPT (but not validated), here's an ffmpeg one-
| liner that may segment based on large-scene-differences:
|
| ffmpeg -i input.mp4 -vf "select='gt(scene,0.4)',showinfo"
| -reset_timestamps 1 -vsync vfr -f segment output%03d.mp4
|
| (Both Claude and ChatGPT are excellent at cooking up ffmpeg
| scripts, but sometimes you need to ask both, or reword the
| prompt a bit to get something that works)
| TomK32 wrote:
| Reminds me of one of my earliest Ruby on Rails apps, some 14
| years ago: Using the flickr api, the whole app white-labelled for
| custom domains etc... never got any traction.
| almog wrote:
| I often use Photos album link sharing option. Have never tried to
| embed in in an existing page, is that the problem it is solving
| at the moment (and relocating the images though I am not sure
| what that entails in terms of usage rights)? I must have missed
| something otherwise.
| orlevco wrote:
| For simply sharing specific photos - GPhotos' native share
| feature is great.
|
| Our product is for people who are looking for a more advanced
| publishing option. Something like a website or portfolio, but
| super quick and easy to create.
|
| Does that make sense?
| a012 wrote:
| If user delete photos on Google side, are they also removed from
| your service? And also what's the retention of user data on your
| service once I cancel my account?
| gabigrin wrote:
| No, the Google Photos integration is uni-directional. We only
| get access to images users pick in the import phase. Removing
| images from MyPhotos is done from the platform itself.
| Regarding retention - we'll delete all your data if you wish to
| cancel your account.
| rishikeshs wrote:
| Hi great idea! I was also trying to build something similar for
| photographers!
|
| But who are your target audience? How is this different from
| sharing a public album link?
| dirkc wrote:
| Well done launching! I had a quick look at the example site
| linked in the footer and I like it.
|
| I've spent way too much time messing around with layouts for
| image galleries with different aspect ratio photos. Looks like
| you're using a masonry layout. Any thoughts on masonry vs
| something like flickr's layout?
| jazzyjackson wrote:
| Sorry to be critical but $15/m before allowing custom domain is
| not a good value IMO, that's right up there with squarespace and
| they'll register the domain for me at that price
|
| This is very much a solution to a problem I have, too, one that
| was taken care of by Google Business sites before they threw it
| in the bin
|
| https://support.google.com/business/answer/https://support.g...
| gabigrin wrote:
| That's a valid point. We're still experimenting with pricing
| and this is great feeback
| yoble wrote:
| My mom is an avid traveler who takes a ton of photos and puts
| them on albums she then shares with the family, and a while ago
| she was asking if I had a solution so she could have a website
| where all her albums are listed. She wanted to not have to share
| a new link each time, and have people easily look up older ones.
|
| I didn't have any easy solution and this looks promising,
| congrats!
|
| A few dealbreaker things I can share:
|
| - she has lots of albums already on gphoto. She'd need to easily
| import them.
|
| - she makes heavy use of the map and text blocks you can add in
| gphoto albums, which makes each album a kind of travel diary. I
| don't get the sense these are supported in your product yet,
| these would be required for her.
|
| - she doesn't have a ton of videos but sometimes she does have a
| few.
|
| - I'd have some concerns about the longevity of your product - if
| she invests time into it she wants to be able to look back at the
| albums in 10-20 years time. Having a convenient way to export the
| albums would be reassuring to me.
|
| - I think she has a few 1000s photos in those albums, so your
| highest tier would be too low for her, if there was a way to buy
| storage that might suit her usage better (though she has a
| hobbyist budget).
|
| It might be technically difficult and you rejected that path
| already, but I'm thinking an ideal way for her would be to keep
| editing her albums in gphotos and sync them to your site, which
| would take care of the longevity concerns and allow her to keep
| using the interface she knows (if you linked directly to the
| pics/vids on google server that would eliminate the cost of
| storing pictures for you, but I assume that's impossible or
| prohibited by google's tos).
|
| Anyway, just sharing my use case in case that's useful but
| congrats on launching and on the good looking product!
| tyrust wrote:
| > she could have a website where all her albums are listed. She
| wanted to not have to share a new link each time, and have
| people easily look up older ones.
|
| Would a static site with a list of links to Google Photos
| albums work? I've done this with my family.
|
| If so, there are ample static website providers out there.
| orlevco wrote:
| First of all, your mom sounds awesome.
|
| Second, thanks for the detailed thoughts!
|
| Easy import of existing albums - We have that! find the album
| on GPhotos, select all and import. Check out how quick it is
| here: https://youtu.be/7gi58SuQ6Rk?si=Wk1wtDoPO5_DYWEV
|
| Map and text blocks - These are currently not supported, but
| we're still considering different use cases and will add
| features to match the ones we focus on soon.
|
| Videos - These are currently not supported but might be added
| soon. They pose some storage and pricing issues but we'll try
| to solve that. Again depending on the use case we end up
| pursuing
|
| Longevity - This is a legitimate concern. Obviously we're
| hoping that this product keeps growing and plan to maintain it.
| Regardless - The north star of this product is to be fast and
| easy to set up. So we hope that the time investment won't be
| that big anyhow.
|
| Pricing and Storage - We've just launched so the pricing plans
| will definitely change over time. We'll take that into account.
|
| Syncing albums - Seamlessly syncing albums to online galleries
| was our dream. Unfortunately Google's API doesn't support that
| (for obvious security and privacy concerns).
|
| We'd love for your mom to give our product a try and will be
| happy to give her a generous period of free usage of the
| premium plan for some of her feedback. If you're interested,
| please reach out at hey@myphotos.site.
|
| Thanks again!
|
| Or
| wisehacker wrote:
| Is it 1000 photos per album? Or is it max 1000 photos
| regardless of number of albums? If it is so, how does
| unlimited albums work in such case?
| orlevco wrote:
| It's currently 1000 photos total that you can divide
| between as many albums you like.
|
| But honestly we've just launched, so pricing and tiers
| might very well change as we discover what users want and
| are willing to pay for.
| dawnerd wrote:
| Been looking for something like this myself and surprisingly
| not that many options out there especially self hosted. All of
| the new ones are trying to chase after AI. Older ones don't
| support object storage.
|
| Using Google for my photos is a no go though so hopefully OP
| will come up with a solution for that.
| orlevco wrote:
| Cool. Can you elaborate a bit on your issue with Google?
| scrollaway wrote:
| I have a different use case in a similar situation.
|
| My girlfriend runs an art studio that does art workshops. She
| uses Google workspace. She and her other teachers use Google
| workspace and have Google photos. They create albums of each
| workshop.
|
| They want to have all the albums in one place so the social
| media person can easily find and use them.
|
| Ideally there should be a way to make a Google photos account
| shared with the rest of the team. Unfortunately Google photos
| is not officially part of Workspace so it's not treated as a
| b2b product.
| welder wrote:
| Reminds me of a 13 year old GitHub project back when Google
| Photos was called Picasa:
|
| https://github.com/alanhamlett/jQuery-Picasa-Gallery
| gabigrin wrote:
| That's cool! thanks for sharing
| dc_rog wrote:
| Super cool.
|
| Typo here "Show your art, keep friends up-to-date or share family
| photos with lode ones."
| mrccc wrote:
| Have you looked into row-based presentation of the images? I
| usually feel like pages with row-based image grids have a higher-
| quality look to them.
|
| See e.g. [1] and [2] for more info and [3] for an implementation.
|
| [1]: https://medium.com/google-design/google-photos-45b714dfbed1
| [2]: https://blog.vjeux.com/2014/image/google-plus-layout-find-
| be... [3]: https://www.npmjs.com/package/react-photo-gallery
| orlevco wrote:
| Super interesting. Thanks for the detailed thoughts and link.
|
| The thinking behind the vertical masonry grid is that by always
| having semi-visible pictures in view it draws the eyes down to
| keep scrolling. If that makes sense.
|
| Will definitely check out the links and consider further.
|
| Or
| podviaznikov wrote:
| very cool. last year I made similar for Apple Photos(but it's on
| pause now).
|
| https://public.photos
| orlevco wrote:
| That's super cool Anton! Your way of uploading the albums is
| really neat. Haven't thought about that.
|
| Do you mind sharing more?
|
| How did that work out for you? Did you encounter any technical
| issues with working with Apple's app? Why did you end up
| pausing the project?
|
| Would love to hear from you,
|
| Or
| podviaznikov wrote:
| I have few other projects like that and this one is one of
| the least popular.
|
| So I paused it.
|
| Also on top of that dealing with photos and videos are not
| cheap and would be harder for me to scale things.
|
| Basically it was a fun little experiment.
| ggandhi wrote:
| This is super cool, just like the www.myphotos.site. Just need
| a nice UI and options to share/hide album.
| october-freedom wrote:
| boycott, divest, sanction
| cryptozeus wrote:
| Good execution but why can't someone just upload the pics on blog
| or Wordpress and do the same ?
| orlevco wrote:
| This is simply much faster and easier than any other way.
|
| Wordpress and other blog sites require quite a lot of effort
| and technical knowledge. We aim to democratize photo sharing
| and let anyone do it quickly and effortlessly.
|
| Does that make sense?
| haarts wrote:
| The one feature I'm always missing is password protection; I
| don't want everyone to see my travel pictures.
| gabigrin wrote:
| Great feedback! Not just great, but 100% made sense to go ahead
| and ship it real quick. It's live! https://youtu.be/qhHHPHDnMTc
|
| (and server-side secure!)
| spike021 wrote:
| Something to consider, adobe Lightroom has a feature coupled with
| adobe portfolio (iirc that's the name of the product) where if
| you use Lightroom you can turn albums into sites.
|
| Similar idea on paper. May be useful to compare how theirs works
| since that could be a competitor of sorts.
| krat0sprakhar wrote:
| Really good execution! I was definitely looking for something
| like this - so great job in working on a real problem.
|
| I think the pricing might be a bit tricky. The free plan allows
| 50 photos whereas the free plan on Flickr allows 1000 photos (and
| no limit on albums IIRC). Can you elaborate a bit on how you plan
| to differentiate?
| orlevco wrote:
| Thank you!
|
| Pricing is indeed something we're still trying to figure out.
|
| Currently we see differentiation as a combination of simplicity
| and elegance/visual quality. Meaning platforms like Flicker are
| too cluttered and outdated in terms of how galleries look.
|
| But in this too we still have a lot to learn and figure out.
| ibzsy wrote:
| really cool
| philsnow wrote:
| One of the things I would want included is any and all metadata /
| tags that Google Photos adds to photos. For instance:
| - inferred/corrected geotags - inferred/corrected time of
| day - names or at least UUIDs or something of people in
| each photo
|
| I almost typed another bullet along the lines of "the top 10
| most-likely features identified in the photo", like if it's a
| photo of a menu or a tower or whatever, then "menu" / "tower",
| but, it's probably encoded as some kind of embedding in some
| space like word2vec or something less ancient, and it wouldn't
| even really be that useful for them to give out the raw embedded
| through the API, would it?
|
| Maybe the last one (faces/people) isn't even possible, if they
| don't reify found people into some actual tag or something, maybe
| you just have to search by person and gphotos just shows you the
| photos with embeddings less than some cutoff cosine distance from
| that person's embedding.
___________________________________________________________________
(page generated 2024-11-28 23:00 UTC)