263 comments

[ 2.2 ms ] story [ 293 ms ] thread
It would be nice to have a few more examples. The only example I see on the homepage is hidden in the video, both hard to see and not always visible.
Yes, I would recommend showing two images (before/after) instead of the video.

You could do more interactive stuff, like the common interface of moving a slider on the "before" image to reveal the "after" image. And that's nice, but it should be supplemental to the two images.

But take this advice with a grain of salt - after all, the video did cause me to spend more time on your site than if two images had immediately answered the question I was curious about.

OP here. Thanks for the advice. I have been considering a few different options. Perhaps I could have a preloaded screenshot with the UI already initialised, so I immediately can show off the functionality? I don't know what is the best approach.
Yeah, I don't want to upload my data to a random website I've never heard of, just to see what the heck it does. You NEED a few examples, otherwise users will just close tab.
I uploaded a screenshot of the front page of hacker news

https://i.imgur.com/aTqN3qA.png

Maybe it's imgur but this looks so bad. A tweet with examples would be better choice if you can't update on the website quickly
having blur set to extreme probably doesn't help but I haven't messed around with it
(comment deleted)
> Yeah,

You sound so cool when you start a sentence with "yeah" for no grammatical reason, as though answering a question that no one asked. I bet you're really laid back and were probably smoking a blunt and exhaling a massive cloud while you typed your comment. Did you personally innovate this conversational technique?

What even is this criticism?

An ad hominem attack based on a single word and your extreme displeasure of it?

[flagged]
Do you think I give a shit about what you think of my superfluous words? Huh?

Do you have anything substantive to say about the comments themselves as opposed to a hyperfixated diatribe based on a single superfluous word? Because everyone and their grandmothers is bored of your “holier than thou” soapbox ramblings about how you perceive random people on the internet.

And just because you clearly don’t get it, the word “even” can be used for emphasis when something you are reacting to is absurd. Crazy thing, English, eh?

[flagged]
Nobody’s ass is hanging out except the whiny pedant who has nothing better to do than to go on ad hominem-filled diatribes based on nothing that is grounded in reality. And who’s comments are clearly absurd enough to get downvoted and flagged. Because they in no way shape or form are about the topic at hand.

I wonder who that person is in this conversation.

Once again, and for the last time, do you have anything to say about the comments themselves, or are you just going to keep breaking the guidelines to get your rocks off on insulting others and their grammar over a single word?

OP here. I understand this instinct. In this case it doesn't actually get transferred, but stays on the client, as stated in the footer: "Image processing is done in the browser. No image data is stored or transmitted."
You should make that much clearer, maybe even by running it in a codesandbox, and submitting that link to HN. I think it might be received better than when marketed as a service (and also, if the process is client-side, why is there a subscription offering?)

Think about Figma. They don't charge for how many drawings you make - that would seem ludicrous, especially since most of the compute power required to use the product is expended during client-side code execution. So instead, Figma charges you for using the app itself, but only once you want to add another collaborator to your private projects (at which point you're obviously happy with the product and probably willing to pay).

If you want to monetize this, I would encourage you to think along those same lines. You are building a product for somebody - but for whom? You can answer that question by letting people use it for free and seeing who likes it the most - marketers? news article editors? - and then building more features to focus the product on their use case.

Don't think about charging by volume - think about getting consistent usage from users who face this problem often, and then figure out how to integrate into their existing workflows. Then charge them money to do that.

100%, I also wonder what this would do to a regular non-screenshot photograph. Could be funky weird or funky cool.
OP here. It's basically Ken-Burns-on-steroids vibes.
ken-burns 1.0

modern ken-burns not only slides the image, but does the 2.5D effects as well. much more interesting than just zoom/pan

> turns screenshots into dramatically angled photos

> much more interesting than just zoom/pan

yes, Ken Burns goes beyond zoom/pan and over into dramatically slanted, at least for what purports to be documentary.

Interesting, would very much like to see an example if you're up for it.
OP here. Valid point! I find it hard to strike a good balance between demonstrating value and cutting to the chase. I wanted it to come off as a utility that lets you take action immediately.
My first reaction was “I can barely tell what this does” so it’s hard to feel motivated to take action at that point.
I had the same response. And beyond the motivation factor, the second response was, "I suppose I could test it myself, but I'm not uploading a picture of my screen to a strange website"
OP here. As per the footer text: "Image processing is done in the browser. No image data is stored or transmitted."
If you've the security consciousness to not upload then there's no way you trust that.
This critique makes no sense. Just monitor the traffic by using a random image if you don't trust the disclaimer?
Monitoring the traffic is consequent to distrusting the site, surely. I too would use an arbitrary image with no PII.
Or just.. close the page and go do something else, which is what I did.
There are some tweets on it, but they are unclickable. I was hoping to expand them and see their examples.
2nd this. Maybe, images with diff dimensions
Great results on the first attempt I did.

When I wanted to export the result in highest available resolution, it asked me to subscribe for $5/month. I only need 1 single screenshot, so I’m not going to subscribe, but I can see this being used by big newspaper companies. Good luck with your project!

OP here. Thanks, that's good to hear. I got journalists at a couple of the big newspaper players here in my country (Norway) to purchase subscriptions, and after seeing these pop up in articles rather often, I felt that I had sold myself short with the $5 monthly fee. One of them (VG.no), is like the 2nd biggest site in Norway, with 70 million visits a month. I guess I'm bad at business.
I think the insight here is you should offer a pay-per-use model too.
Took me 2 scrolls to find an image probably generated by your tool.

https://www.tek.no/nyheter/nyhet/i/KnWn0G/ai-flause-for-goog...

You probably should make a personal account which limits the amount of images you can generate a month (10?) and a business account which offers unlimited generation but for a steep price.

Or maybe allow free usage with an added watermark. So that people can try it out on their own screenshot, eventually distribute them internally, and pay when they need to remove the watermark for public distribution.
I think that's how the SD basic tool works.
$5/yr is too much for the casual user and too little for big newspapersq
I made a screenshot using this and it's pretty cool.
I'm sure you put a ton of work into this, so it's really unfortunate that I can barely tell what this really does or why we should care.

Also the "Ready to go pro?" thing seems very rushed and presumptuous. Like, I'm not even sure I'm ready to go "Free" yet

> so it's really unfortunate that I can barely tell what this really does or why we should care

It takes an image as input and produces a "dramatic" copy of that image in which "dramatic" is defined as being angled away from the viewer on two planes and having a blur somewhat consistent with it having been photographed with a shallow depth of field.

(comment deleted)
I'm surprised to read this — while the examples could be more cleanly presented, I felt like the title for this post and the initial gif on landing show what this does reasonably clearly. I do see what you mean about the PRO part.
Why not skew, mask and blur with photoshop in 10 seconds?
you mean aside from the fact that it requires downloading a huge multi-purpose program that starts at $21/mo which is built for significantly more than just those features and the user must use a lot more brain space for what they want to be a simple task?
Or gimp. "Photoshop" is the all encompassing term for these editors. This task is actually quite trivial, but there is a bit of a learning curve for someone who knows absolutely nothing about photo editing.
For a Linux user, you can already build such a system yourself quite trivially by using PHP's ImageMagick extension to modify the uploaded image file.
I wonder can ImageMagick give you the "progressive linear blur" (or whatever you want to call it) that emulates depth of field?

Might as well ray trace it....

Do it for free with Photopea.

And this would be just about one of the easiest "Photoshop tutorials" there is to follow.

Photopea is nice, but their website is not friendly. I could see why people would prefer to pay and have a simple website to use.
(comment deleted)
That was my thought as well and you can go even further by just making it a script. Which begs the question: why is this a subscription service and not the script.
I can see how this is useful for laypeople. Those with a DIY attitude might resort to popular image editing tools. My choice would be trying to achieve this with Imagemagick. Fred Weinhaus' script http://www.fmwconcepts.com/imagemagick/skew/index.php and maybe adding a blur later.
In Gimp it's Tools, Transform Tools, 3D Transform.

This is a good idea to add visual novelty to images, especially plain ones.

The problem is the moat is very low: as low as 1 30-second YouTube video on using the transform tool in the case of Gimp.

Doesn't seem very defensible - but maybe if you add versioning & various other features (links for images, etc.) it could be.

One can make so many fun things as client-side web tools by compiling ImageMagick to Wasm!

One thing I played around with because I needed it myself was turning pictures into things that look like they've been scanned. Made for documents, but I suppose one could use it for screenshots as well. :) https://photocopy.fuglede.dk

This is a cool idea. Wish website showed a bit more info and examples but I like what it d oes. Content creators could use this.
OP here. Thanks! And yeah, I will try to improve on the landing page.
What is this paid subscription to be able to rotate the screenshot a certain way?
Yes, you pay for the "your mother who takes screenshots by photographing her screen" effect.
Ooh, I used to take 21MP photographs of LCDs as a cool effect. I really loved doing that. It looked so aesthetic.
This looks neat indeed :)

I have no idea how hard it would be, but this would be great as an effect for Final Cut, Da Vinci, etc. I don’t think the video world has a plug-in standard like the photo kinda does (eventually lots of apps implemented Photoshop's API).

OP here. I'm fantasising about making a video version of this app, to generate some very cool documentary-style Ken Burns-shit imagery.
A version that pops out the video on the site could be cool (like, takes in a screenshot but then tilts and zooms it as a video clip). Very “nightly news.”
I clicked the link because it sounded interesting and maybe useful. But then I wasn't quite sure what it did.

I wanted a before and after photo. It's a simple concept. The landing page should be equally simple.

OP here. A few others have made this point, and it's glaringly obvious to me in retrospect. Guess I have my work cut out for me.
There’s a gif on the landing page that shows what it does pretty much immediately
That's pretty much what the animated image is. It shows a screenshot before, the tool being used, and the final image as part of a tweet.
(comment deleted)
Based on the GIF, I thought it created an animated video.

Even when the .PNG downloaded I thought for sure it’d be an animated PNG.

If I’m doing some content creation, I probably already have an image editor, in which case I can create this effect myself or would prefer an integrated plugin to do it.

Motion graphics is much harder, and there’s more demand there to add some sparkle to a static image. OP, have you considered that angle?

OP here. Sorry to disappoint you with the lack of animation. I have definitely entertained the idea of creating a video variant of this app. I fear it will remain a pipe dream due to the demands of my day job and family life.
"Charged monthly at only $5 for individuals or $3 per person for teams of 3 and up. Cancel anytime." - excellent tool, terrible business model.

No on wants to pay $5/month for a utility.

(comment deleted)
OP here. How would you monetize it?
Yeah, subscription for the big players (and raise your subscription price), then $1 a pop for a one-off.
What if a corp signs up as a consumer and uses $1 pops?
Mr. TotallyNotCorporate which makes 20 orders a day? Totally not a corp.

Just add "more than N screenshots a day are considered corporate usage with an according plan", ask to upgrade, ban.

Just let them give you $20 a day.
I would consider the value to the consumer and use a value based pricing model.

The value you're offering is overcoming a few challenges.

First, the technical challenges of designing these graphics yourself - you have to pay for potentially expensive software suites (Photoshop, etc.) to get this kind of work done, and then you have to learn how-to do it. There is a financial and educational curve to climb there.

Then the second challenge is time. It takes time to sit and fiddle with a complex piece of software to make a screenshot do what your utility does.

So your value is you provide a single piece of software that does one job very well, and it's near instant and requires little to no learning curve whatsoever.

Further, you need to consider your own personal objectives with this software. I'm not in this field, so I do not have the foresight to see the potential growth with this software, therefore from my perspective it's a cute tool that does one thing. You might know different and can see potentially big markets.

Anyway. I would likely price this as a $15 one time purchase at the non-commercial level (targeting everyday Joe Blogs) with one year of upgrades, and around $50/year/user at the commercial end.

It really depends on the market and the user's pain points, Jobs To be Done, etc.

Maybe a couple of freebies (with watermarks with link to your site for viral growth and license for personal use), then upsell with credits. You mentioned newspapers - I think you could charge a bit more for a full commercial license for the outputs - but it is fairly trivial for a graphic designer to replicate this in photoshop in 2 mins so there is probably a limit to the max fee.

Subscriptions are good for recurring revenue but maybe offer that as one option, not the only option? A lot of people drop off at that point.

I used to do these tilted 3D dof screenshots in After Effects a lot for videos - that is a use case I would pay for, if it’s quick enough - it was always a bit time consuming in AE and also a lot of people need videos but not everyone can use AE.

For my own app https://screenrun.app/ that also produces outputs, I was thinking credits: eg $5 gives you 50 screenshots (or videos in my case). No subscription and people get debited only when they use the tool. Would be nice to add referrals so they could earn credits when others sign up?
Yeah, I was initially interested, but then I saw that it was a subscription and I just stopped looking.
While the tool is extremely simple to replace with something more low level. I think there is no better way than what he chose to actually monetize something like this. Consider the target markets.

Individual bloggers. Too cheap to pay sub. Also too cheap to pay a $10 onetime.

Hackers. Would happily spend whole day implementing the same with a less polished npm package and would never pay for anything anyway.

Professional designers. Can get the same or even better results in the tools they already use with both hands tied behind their back. They don’t need this even if it’s free.

Now to the last and only viable market. Journalists who are lacking tech skills and no designer to back them up. They don’t care how simple the tech behind is, as long as it gives them value. It’s just cheap enough that they don’t have to blink and the recurring revenue from these will offset any non-existent $10 one-off purchases from individuals.

I would reconsider this pricing model. Flat monthly fee with unlimited calls/runs may not be pleasant for you.
OP here. I don't do any cloud computing to deliver the service. It is essentially a pure JavaScript-application running on the client.
I have a similar tool that I am working on that is basically javascript on an html page.

There are probably other ways to add value for paying users so they are less likely to just save page as:

Integrating the JS into a WordPress plugin to generate screenshots automatically for posting to social media when a post publishes.

Zapier integration to help people automate these screenshots in their workflows.

then what's stopping me from sandboxing this and .. not paying anything?
Goodwill. And the fact that people can do any number of other things to get the same effect without paying. What are you really asking?
[flagged]
(comment deleted)
Why do you need to use a snide tone to convey your point? Just make it in plain language, don’t be rude.
Yeah fair, I'm one one today. I genuinely thought this was a joke at first. But you're right, I should just put my money where my mouth is and make a github pages or similar that replicates this functionality as something that you can download and keep forever, how hard can it be?

It's utterly disgraceful to charge money for something like this as a service, but that doesn't excuse my conduct in response. The implications of validating this sort of rent collecting are insidious and serve. We really need to re-evaluate what our definition of 'value' is as a society.

Not everyone shares your values, and that's OK
I seem to recall a similar discussion yesterday. I agree with you but I think I'll sit this one out :)
I may have made some bad choices in the past 48hours. Yr def making the wise choice. <3
What people are willing to pay for it is the value of it, I guess. Capitalism :shrug:
I can think of a number of reasons that make this valuable:

If someone just wants an image like this and doesn't even know what css is, it's cheaper to pay 5 bucks than to even consider hiring someone to do this.

If someone does know how this is done... Why would they purchase the service?

Also, what if someone wants this but it's not for a website? They make pamphlets or something.

My only complaint is that from the example I thought it would return an animation, not a still image.

> It's utterly disgraceful to charge money for something like this as a service

Why?

> We really need to re-evaluate what our definition of 'value' is as a society.

If this saves me about half an hour a year it's valuable because it's cheaper. If it's easy to make, other people will come along to profit at lower price points.

Are you new here? Perhaps I can introduce you to one of our heros, patio11 who made an entire business out of making bingo cards. I don't particularly think this is a great business but my gosh, this is the right audience to "show it off"

  For a Linux user, you can already build such a system yourself quite trivially by getting an FTP account, mounting it locally with curlftpfs, and then using SVN or CVS on the mounted filesystem
Reference - https://news.ycombinator.com/item?id=9224
Wrap it up as a service and I'll give it a try; not a fan of writing code
(comment deleted)
Error on the landing page:

> Gorgeous screenshots is just a click away...

should be "Gorgeous screenshots are just a click away...".

I'd pay for one-time pro features. I don't want a subscription, but I'm willing to pay to use this service a few times a year maybe. There are probably more like me.

Maybe sell 10 pro edits or something like that?

I think this would be a tough one. FWIW, I think your question highlights a great example of why so many companies don't offer the pricing you want, even though it's probably true that the majority of users would prefer this pricing.

My bet is that selling "10 pro edits or something like that" would nearly completely cannibalize any subscription sales. Thus, I think OP is probably likely to make a lot more money by only offering a subscription than by offering one-off pricing that kills subscription sales.

I see where you're coming from, and I don't really have the market experience to argue with you.

In my case 10 pro edits would be way too much, it could expire at the end of the year too, to make it even more profitable. First of all it would be more expensive than one month of subscription, and second of all they would expire.

But a person like me who sucks at graphic design would still be willing to fork over like 10 bucks for 10 pro edits even if I only use a few of them.

Because once that rare moment comes where I want this, I really want it!

(comment deleted)
I have no use for this. Very cool though. Reminds me of local news stories about the dangers of the internet
Interestingly, I REALLY want the exact OPPOSITE of this.

Many times it's just so frustratingly difficult to take a screenshot of something on a computer screen and send it over one of the several mobile messaging apps that don't have a good desktop client, so I end up just taking a picture of the physical computer screen with my mobile phone. It looks like shit but I'm lazy. I really wish a piece of software could recognize what I just took a picture of and then pull the actual data from that portion of the screen, send the screen data over bluetooth or Wi-Fi or whatever to the phone, and "fix" the screenshot.

(comment deleted)
On Apple devices ”Universal Clipboard” (https://support.apple.com/en-us/HT209460) works fairly well for that (’fairly’ because, in my experience, it doesn’t always work in one go)
'KDE connect' has a similar shared clipboard, also very low friction local network file sharing IME, works on Android with either Windows or Linux (and maybe other systems too?).
I had a similar problem a decade ago. Had a lot of paper bureaucracy in my life, didn't have a scanner on me at all times obviously. So I wrote an app that undoes the perspective: https://github.com/akalenuk/Docam
How’s the SaaS model working?
OP here. I'm not quitting my day job anytime soon. It's a grind to get the word out in the first place, for one. The SaaS model covers some of the cost of running it, but it isn't a viable product on it's own just yet. I have other projects that generate income to cover cloud services, accounting costs and other fees.
I tried to use the tool, but it's very laggy/slow. Is it server-side? Why not just HTML5?
I think it's client side. At least his other project showcasted on the linked persille.no website is.
OP here. Sorry to hear it. What browser and OS are you on?
Brave, on a 4K monitor
Why would this be a subscription? It could just be a standalone app for $10 bucks and you'll probably get more lifetime revenue that way compared to having a subscription, because in a year more people will buy for $10 than will subscribe and stay subscribed for long enough to overcome that revenue.
> Why would this be a subscription?

Because more monies.

Could also make it hard to unsubscribe. Like swedish insurance, where it needs 3+ hours waiting on the line to get to some supporter who can unsubscribe.

No, it’s literally less money.
> Because more monies.

See how many people are saying the pricing model is wrong? They won't get hooked on a sub because they'll never sign up.

Right. I (reluctantly) accept subscriptions if there is something the company does for me every month - in this case it's a javascript app that runs in my browser. There is no countervalue per month, making it unattractive.
With all due and sincere respect to mikaelaast's effort and ambition, I can't believe we're at a point where the implementation of an image filter can ask for a automatically renewing $5 per month subscription and nobody blinks.

We used to blink, didn't we?

I didn't just blink; I immediately ignored.

And yes - same comment; awesome tool, but lol - not worth paying for.

If they really want to monatize: Enforce account sign-up, go on a token system for each use, and give a few free for trial maybe.

(comment deleted)
and nobody blinks.

There are a whole bunch of repetitive comments about this already and it's not, by itself, all that on-topic in a Show HN, especially with the added meta.

At this point I believe that this is only a snowflake on an enormous iceberg of opportunities that we are missing, because “come on, everyone could open a shell and <insert your convoluted pipeline here>”. It’s probably a trillions dollars industry that doesn’t exist due to professional deformation and being conditioned to sit and work for someone else’s profit.
You are not the target market.
My 100 Croatian lipa :

This feels like something that's too expensive for a random individual (I've become allergic over time to simple things that insist on subscription as opposed to pay once). But too cheap for enterprise use.

Consider putting together few more simple tools, build a quick pipeline, and offer per seat and enterprise licensing.

(comment deleted)
(comment deleted)
I agree, I would also try to market this is as one tool (“Dramatic Screenshot”) of a larger “Screenshot Toolbox” or sth.
It's 13 Croatian Euro cent now ;)
[flagged]
This is why we can't have nice things

(Note: I don't know OP, and have never used their product)

I mean, the fact that it’s not properly protected doesn’t really give you the moral right if “don’t want to pay”.
If the author was that serious about this as a business they would've done it server side -- or better yet sold the unburdened source code at a one-time price.

If you are running someone else's code on your computer you are entitled to change it, morally. The fact that most other programs make this hard to meaningfully do is another problem.

Upside to you is some karma points. Downside is harming their business, right there on the launch post. Why? Just quietly notify them offline.
Eh, from the rest of the thread I don't think these 133t HN readers are the target audience anyway.
I appreciate the reveal. I come to HN to see things exactly like this. And if someone is going to “show HN”, well hacker is right in the title of the site so expect people are interested in that kind of thing so best take minimal steps to secure your stuff.
“Business”? This is a weekend toy project, and good for them for trying to explore monetization as part of the exercise.

But reverse engineering the implementation, even trivially, is more intellectually interesting than the effort itself. It’s good content for HN and a really valuable beginner’s lesson for the author.

A bit unfair of you to just plainly state how to circumvent their protection in the comments.
The name of this website is Hacker News.

That said it’d be unethical to use this to avoid paying.

There _is_ no protection. the source is publicly accessible.
The conditional in the code _is_ a protection. It is not fair of you to consider it non-existent just because _you_ can easily bypass it.

Using your argument, it would be fair to circumvent any protection as long as a specialist can trivially bypass it? Remember that 99% of internet users have no idea what "Show page source" actually does, much less what "javascript" is. You are the top 1 percentile power user here.

Arguably a lot of things are not strongly protected per se but rely on specialists not trying to actively trick the system, much less posting their techniques on public forums.

> Arguably a lot of things are not strongly protected per se but rely on specialists not trying to actively trick the system

If a system has a vulnerability, it is the system's fault, not the specialist's. I'm a bit sensitive to this line of argument because it and arguments like it have been used in the past to try and demonize the efforts of security researchers. However, I also believe we should have good norms around responsible disclosure.

> much less posting their techniques on public forums.

Perhaps it could've been more responsibly disclosed to the OP, agreed. However, the public internet is a public forum. OP posted his code there, and it was found. Any number of web crawlers have probably already found it. Not the least painful way to learn this lesson, but hopefully it was learned.

(comment deleted)