4 ms·

Show HN: X-frame – embed any content with a server-rendered super iframe

4 points by graderjs 5y ago

bryanrasmussen 5y ago
I think you should put it to the front page and not the purchase page, because on the purchase page you don't have any clue what it's for or what the use cases you have envisioned for the technology. Once I did look at it I thought that might be useful.
- graderjs 5y ago
  I just want to thank you for that but also say that I like it how it is. I'm happy with it and like it, I think it's perfect for what I want and good, but I also get your point. I love the simple style and I also think I can improve marketing and polish. But I won't do it for this one because it's just a silly demo. Mentally, I think this is better than my main page. I tried to keep it simple with the three options: [free demo ~ try now] [basic purchase] [expert purchase] just giving an overview of features, and then once you click 'try now', you can use it right away to get a better idea. I think it's good :)
  - graderjs 5y ago
    Alright, I've thought about this some more and I think I want to repost in future to make it really obvious the value to developers. Here's my current todo list: - [ ] solve the meta bug - [ ] ensure Flash works but disable secure doc viewer (with a note 'available in Pro') - [ ] use docker single file scale server with simple AUTH - [ ] create a "code <-> result" demo where people can see how to use an iframe to embed this. - [ ] set options (like max-tabs, display Chrome UI, etc, allow sound, allow devtools, allow flash, allow chat, allow cobrowsing, allow sound) via environment variable - [ ] basic co-browsing indicator of other users (circle highlights on pointer) - [ ] make demos live for only 5 minutes (docker stop / kill) - [ ] show an "API integration" where you can "fully script" the browser using puppeteer etc - [ ] new static landing page. repost "X-FRAME" - [ ] make it also work via a custom element (x-frame sandbox="co-browsing") etc - [ ] show API for "record of interactions" - [ ] show API for "record of screenshots" - [ ] also custom VF options like: stop frames, start frames, resize, new tab, close tab, open devtools. - [ ] marketing copy: "x-frame - unrestricted embedding; a super iframe; embed any web content with a super iframe; scriptable remote browser isolation; drop-in full featured web-proxy right in your web client" - [ ] expose the puppeteer API right on the x-frame itself or in the JavaScript, so like from the client you can call: "selectorGet" or whatever the fuck and you get a result. Nice! - [ ] enable option to kill browser when a client has disconnected (no websocket, for more than 60 seconds). this is all about pitching developers to ensure they like it. - include minor section on caveats but be really specific like: "when you use the built-in remote devtools front-end (screenshot) right now it only works on Chrome reliably, and Firefox to some extent, while Safari has multiple bugs which make many DevTools features unusable...but cross browser devtools front end is actively being worked on and the Chrome team support its eventual inclusion in a future version of chrome." - also -- pricing: I think "pay per session" is a good idea. Or "Pay per total booked time." is a good idea. - how about i also reach out to people who have reached out to me before about this and tell them about the HN post: B**, D**, S**, R**, A**, N**, M***, etc
graderjs 5y ago
Right now this demo only has 3 tiny servers (singapore, iowa and n virginia). If it becomes popular I'll add more tiny servers. Manual scaling behind a client-side GitHub static site hosted "load balancer" that picks the server with the closest roundtrip. Also note that these demo servers are shared, so you might be fighting another use for control of the x-frame in real time. ha. I just thought I'd try this approach.
bryanrasmussen 5y ago
I think one of the problems with this, which I've ran into with tools I've built in the past, is that there are a lot of potential use cases but to make one of those use cases complete would take a lot of work - for example co-browsing, probably you want that as a part of something else. So probably you should identify types of companies that would benefit from co-browsing in their products, find companies that are successful enough to pay you for the X-frame licensing, and then try to market to them to integrate your product in theirs.
- graderjs 5y ago
  But I've already got co browsing working. It really doesn't take that much to make it work.... I mean when I was developing the co browsing feature I think it was around a week to 10 days of work to get multiple clients set up but really the architecture I chose initially made that extension pretty easy. And then of course there's things you can add on top of that like having a real Time chat which I've already got. And the next thing I'm going to add is just indicating where the people's curses are so when multiple people are driving it, everyone knows where the others are. Also it wouldn't be hard to add a flag to indicate that only one person is the leader driver and the other people are sort of just observing. Really the co browsing thing is not that hard to make work I've already got it happening. But you seem to know about this so I mean please keep going with more examples or discussion about this I find that really interesting and valuable so if you have anything more to say please keep going. thank you very much.
  - bryanrasmussen 5y ago
    I mean one thing I would expect that I would want co-browsing in is: 1. some sort of remote education app 2. some sort of remote debugging, testing app 3. onboarding in another app. Those apps will have more things essential to them then co-browsing. Co-browsing is a really nice to have, maybe even killer feature to make such an app stand out from their competition. But in the end I would think it was still a feature of the app. Onboarding in another app, I guess they would just do that as needed with someone. Really a sort of educational feature. Let's say for example you are in Denmark and you have to use an ESDH solution for your government job (electronic case documentation handling solution [case in Denmark is sag]), they gave you some classes you went and you don't understand some of the things you need to do, or two you were sick for those classes and the company doesn't want to do a single class for you. Then co-browsing is pretty great, but even in that case where they probably don't care to have to full control over experience you should probably try to market directly to them because unlikely they will see your stuff on HN and think THAT IS WHAT WE NEED! As an anecdote of people not seeing the point of what something is and having to market directly to them and convince them, I was once working on a multiple media generating product - generate your website pdf email help files etc. media from single source - reuse documents between media easily. At one demo for a sales person and a technical guy I showed how you could create a branding experience in one media, and also stuff like contact details for your company, and it was automatically inherited across the different media etc. without needing to do anything new for each media and these guys were incapable of understanding the purpose of this or why anyone would find it at all useful. What I'm saying is the co-browsing is cool, but to sell it you are probably going to have to make real specific use cases for people that hammer into their heads why they should want this because what is obvious to you just will not be obvious to everybody.

4 points by graderjs 5y ago

I think you should put it to the front page and not the purchase page, because on the purchase page you don't have any clue what it's for or what the use cases you have envisioned for the technology. Once I did look at it I thought that might be useful.

graderjs 5y ago
I just want to thank you for that but also say that I like it how it is. I'm happy with it and like it, I think it's perfect for what I want and good, but I also get your point. I love the simple style and I also think I can improve marketing and polish. But I won't do it for this one because it's just a silly demo. Mentally, I think this is better than my main page. I tried to keep it simple with the three options: [free demo ~ try now] [basic purchase] [expert purchase] just giving an overview of features, and then once you click 'try now', you can use it right away to get a better idea. I think it's good :)
- graderjs 5y ago
  Alright, I've thought about this some more and I think I want to repost in future to make it really obvious the value to developers. Here's my current todo list: - [ ] solve the meta bug - [ ] ensure Flash works but disable secure doc viewer (with a note 'available in Pro') - [ ] use docker single file scale server with simple AUTH - [ ] create a "code <-> result" demo where people can see how to use an iframe to embed this. - [ ] set options (like max-tabs, display Chrome UI, etc, allow sound, allow devtools, allow flash, allow chat, allow cobrowsing, allow sound) via environment variable - [ ] basic co-browsing indicator of other users (circle highlights on pointer) - [ ] make demos live for only 5 minutes (docker stop / kill) - [ ] show an "API integration" where you can "fully script" the browser using puppeteer etc - [ ] new static landing page. repost "X-FRAME" - [ ] make it also work via a custom element (x-frame sandbox="co-browsing") etc - [ ] show API for "record of interactions" - [ ] show API for "record of screenshots" - [ ] also custom VF options like: stop frames, start frames, resize, new tab, close tab, open devtools. - [ ] marketing copy: "x-frame - unrestricted embedding; a super iframe; embed any web content with a super iframe; scriptable remote browser isolation; drop-in full featured web-proxy right in your web client" - [ ] expose the puppeteer API right on the x-frame itself or in the JavaScript, so like from the client you can call: "selectorGet" or whatever the fuck and you get a result. Nice! - [ ] enable option to kill browser when a client has disconnected (no websocket, for more than 60 seconds). this is all about pitching developers to ensure they like it. - include minor section on caveats but be really specific like: "when you use the built-in remote devtools front-end (screenshot) right now it only works on Chrome reliably, and Firefox to some extent, while Safari has multiple bugs which make many DevTools features unusable...but cross browser devtools front end is actively being worked on and the Chrome team support its eventual inclusion in a future version of chrome." - also -- pricing: I think "pay per session" is a good idea. Or "Pay per total booked time." is a good idea. - how about i also reach out to people who have reached out to me before about this and tell them about the HN post: B**, D**, S**, R**, A**, N**, M***, etc

graderjs 5y ago

Right now this demo only has 3 tiny servers (singapore, iowa and n virginia). If it becomes popular I'll add more tiny servers. Manual scaling behind a client-side GitHub static site hosted "load balancer" that picks the server with the closest roundtrip. Also note that these demo servers are shared, so you might be fighting another use for control of the x-frame in real time. ha. I just thought I'd try this approach.

bryanrasmussen 5y ago

I think one of the problems with this, which I've ran into with tools I've built in the past, is that there are a lot of potential use cases but to make one of those use cases complete would take a lot of work - for example co-browsing, probably you want that as a part of something else. So probably you should identify types of companies that would benefit from co-browsing in their products, find companies that are successful enough to pay you for the X-frame licensing, and then try to market to them to integrate your product in theirs.

graderjs 5y ago
But I've already got co browsing working. It really doesn't take that much to make it work.... I mean when I was developing the co browsing feature I think it was around a week to 10 days of work to get multiple clients set up but really the architecture I chose initially made that extension pretty easy. And then of course there's things you can add on top of that like having a real Time chat which I've already got. And the next thing I'm going to add is just indicating where the people's curses are so when multiple people are driving it, everyone knows where the others are. Also it wouldn't be hard to add a flag to indicate that only one person is the leader driver and the other people are sort of just observing. Really the co browsing thing is not that hard to make work I've already got it happening. But you seem to know about this so I mean please keep going with more examples or discussion about this I find that really interesting and valuable so if you have anything more to say please keep going. thank you very much.
- bryanrasmussen 5y ago
  I mean one thing I would expect that I would want co-browsing in is: 1. some sort of remote education app 2. some sort of remote debugging, testing app 3. onboarding in another app. Those apps will have more things essential to them then co-browsing. Co-browsing is a really nice to have, maybe even killer feature to make such an app stand out from their competition. But in the end I would think it was still a feature of the app. Onboarding in another app, I guess they would just do that as needed with someone. Really a sort of educational feature. Let's say for example you are in Denmark and you have to use an ESDH solution for your government job (electronic case documentation handling solution [case in Denmark is sag]), they gave you some classes you went and you don't understand some of the things you need to do, or two you were sick for those classes and the company doesn't want to do a single class for you. Then co-browsing is pretty great, but even in that case where they probably don't care to have to full control over experience you should probably try to market directly to them because unlikely they will see your stuff on HN and think THAT IS WHAT WE NEED! As an anecdote of people not seeing the point of what something is and having to market directly to them and convince them, I was once working on a multiple media generating product - generate your website pdf email help files etc. media from single source - reuse documents between media easily. At one demo for a sales person and a technical guy I showed how you could create a branding experience in one media, and also stuff like contact details for your company, and it was automatically inherited across the different media etc. without needing to do anything new for each media and these guys were incapable of understanding the purpose of this or why anyone would find it at all useful. What I'm saying is the co-browsing is cool, but to sell it you are probably going to have to make real specific use cases for people that hammer into their heads why they should want this because what is obvious to you just will not be obvious to everybody.