💾 Archived View for gemi.dev › gemini-mailing-list › 001000.gmi captured on 2024-05-12 at 16:20:56. Gemini links have been rewritten to link to archived content
⬅️ Previous capture (2023-12-28)
-=-=-=-=-=-=-
Hi everyone, I am new to Gemini space and I have created gemini library in GO: https://github.com/kulak/gemini Its main advantage is that it follows patterns established in [net/http] package, which makes it easy for people to start writing Gemini servers. I also adapted my favorite httprouter package to this library: https://github.com/kulak/geminirouter The code seems to work well and I have a small test site: gemini://www.argfac.com The website is not of much value as it simply shows Gemini and Titan based chat setup that I have tested only on Lagrange. I used the following resources: - https://gitlab.com/gemini-specification/protocol/-/blob/master/specification.gmi - https://communitywiki.org/wiki/Titan I think I saw somewhere a reference to another "update" resource protocol, but I just could not find it. Titan documentation was a bit on a hard to consume side and I actually had to dig Lagrange network level code to figure out what Titan is supposed to do. I think the following are core promises of Gemini that make it great: 1. no user tracking 2. demonetization as partial result of #1 3. no adds, because it requires user to click on image. 4. true focus on content delivery and not the presentation. The improvement in usability due to ability to apply "stylesheets" at browser level and not at creator level. This itself has huge positive impact on ability to consume information. However, I have some criticism that comes with positives. Gemini as of today will not scale in interactive usability, because user interaction is severely limited. There is a statement that Gemini is only for consumption. However, we have to be realistic that for usability of consumption to stay high, user cannot be asked to live Gemini browser for HTTPS powered web browser to submit a post. That breaks user flow and makes Gemini not worth it. Why to come back to Gemini if I just used standard browser to do something and in the process I compromised all the advantages listed above? Titan protocol is a good step in the process and perhaps is the only required step. I am not sure. We have to remember about need to support multiple screen types from desktop to mobile and it is tough to create something acceptable and generic for user input under the restrictions Gemini protocol put in place. I am hoping people here come up with something :) I also would like to raise people's awareness, that if the network to grow and get any close to big, people shall expect that network would promote payments models, because it kills advertisements so efficiently. I think it is a good thing. I want to consume a product and not to be a product. If I am a product, then UI tends to focus on makings adds prominent and waiting my time. That goes back to value of Gemini and consequence of its proposition is non-add based models. I hope.
On Mon, Aug 30, 2021 at 7:43 AM Sergei Gnezdov 'sergei.gnezdov at gmail.com' <gemproj+sergei.gnezdov=gmail.com@suckless.anonaddy.com> wrote: > > Gemini as of today will not scale in interactive usability, because user > interaction is severely limited. There is a statement that Gemini is > only for consumption. However, we have to be realistic that for > usability of consumption to stay high, user cannot be asked to live > Gemini browser for HTTPS powered web browser to submit a post. That > breaks user flow and makes Gemini not worth it. Why to come back to > Gemini if I just used standard browser to do something and in the > process I compromised all the advantages listed above? > I know this is not very progressive to say it, but why does it need to scale? Why does everything always have to appeal to everyone? Not to mention the well-discussed issue as to whether content *needs* to be updated through a browser. My stance on this is pretty much: you have to learn things to use things. And using sftp to upload a file as a bare minimum requirement is not hard. And it uses a protocol made for file transfer! You can even transport images with it! </sarcasm> In all seriousness, there are multiple real life examples of this. If you want to learn to drive a car you need to consider learning how to drive. Who'd think that just because you know how to ride a bike, someone tailors your car to work like a bike, you know? geminispace provides a nice retreat from the complexities of the clearweb and it also makes you get back to tools that were around before everyone decided that literally everything needs to live in a browser. It doesn't mean that it's uncomfortable to use, but as skyjake once put it on gemini:// station.martinrue.com: there's a reason text editors are there, because text editing is hard. Adding that to a browser is tough. If you make titan:// mandatory in geminispace, most browsers will instead to turn to the shell where you have immediate access to nano, vim and emacs, and what have you gained then? I know I am ranting. But it's been discussed. geminispace is feature complete from a protocol end. And if it doesn't grow big, that's probably ok. Because the people that are in geminispace are great as it is :) > > Titan protocol is a good step in the process and perhaps is the only > required step. I am not sure. We have to remember about need to > support multiple screen types from desktop to mobile and it is tough to > create something acceptable and generic for user input under the > restrictions Gemini protocol put in place. I am hoping people here come > up with something :) > > I also would like to raise people's awareness, that if the network to > grow and get any close to big, people shall expect that network would > promote payments models, because it kills advertisements so > efficiently. I think it is a good thing. I want to consume a product > and not to be a product. If I am a product, then UI tends to focus on > makings adds prominent and waiting my time. That goes back to value of > Gemini and consequence of its proposition is non-add based models. I hope. > > > >
I think that we could add to extend the 1x status codes to support form inputs and secure form inputs. Each field could have validation rules and hints to specify to the client that a field’s value should not be echoed. The size of the each submitted field would also need to be sent to the server so that the server knows when input ends. With this in place, we have something that at least covers text data submission. As an alternative, we could either merge the Gemini and Titan protocols or keep them separate but refer to Titan in the Gemini spec for “advanced” input. Regarding payments, how would 3D-Secure work for traditional card payments? Maybe we could also look into digital wallets? Lastly, I think section 1.6 of the Gemini FAQ is something that may be worth taking a look at again: https://gemini.circumlunar.space/docs/faq.gmi. Gemini never was intended to replace the web. I’m completely fine with these suggestions I made being completely dismissed. Some problems are just better suited for another protocol to handle. What so you all think? I see an opportunity for another protocol to be developed instead that builds upon the foundations of Gemini but actually does try to replace the web. P.S. Thank you for making for creating a Gemini library! > On 30 Aug 2021, at 06:43, Sergei Gnezdov <sergei.gnezdov@gmail.com> wrote: > Hi everyone, > > I am new to Gemini space and I have created gemini library in GO: > > https://github.com/kulak/gemini > > Its main advantage is that it follows patterns established in [net/http] package, which makes it easy for people to start writing Gemini servers. > > I also adapted my favorite httprouter package to this library: > > https://github.com/kulak/geminirouter > > The code seems to work well and I have a small test site: gemini://www.argfac.com The website is not of much value as it simply shows Gemini and Titan based chat setup that I have tested only on Lagrange. > > I used the following resources: > > - https://gitlab.com/gemini-specification/protocol/-/blob/master/specification.gmi > - https://communitywiki.org/wiki/Titan > > I think I saw somewhere a reference to another "update" resource protocol, but I just could not find it. Titan documentation was a bit on a hard to consume side and I actually had to dig Lagrange network level code to figure out what Titan is supposed to do. > > I think the following are core promises of Gemini that make it great: > > 1. no user tracking > 2. demonetization as partial result of #1 > 3. no adds, because it requires user to click on image. > 4. true focus on content delivery and not the presentation. The improvement in usability due to ability to apply "stylesheets" at browser level and not at creator level. This itself has huge positive impact on ability to consume information. > > However, I have some criticism that comes with positives. > > Gemini as of today will not scale in interactive usability, because user interaction is severely limited. There is a statement that Gemini is only for consumption. However, we have to be realistic that for usability of consumption to stay high, user cannot be asked to live Gemini browser for HTTPS powered web browser to submit a post. That breaks user flow and makes Gemini not worth it. Why to come back to Gemini if I just used standard browser to do something and in the process I compromised all the advantages listed above? > > Titan protocol is a good step in the process and perhaps is the only required step. I am not sure. We have to remember about need to support multiple screen types from desktop to mobile and it is tough to create something acceptable and generic for user input under the restrictions Gemini protocol put in place. I am hoping people here come up with something :) > > I also would like to raise people's awareness, that if the network to grow and get any close to big, people shall expect that network would promote payments models, because it kills advertisements so efficiently. I think it is a good thing. I want to consume a product and not to be a product. If I am a product, then UI tends to focus on makings adds prominent and waiting my time. That goes back to value of Gemini and consequence of its proposition is non-add based models. I hope.
That's dedication! I presume that was a lot of work. Have you talked to the person who invented titan:// ? I think gemini:// transjovian.org/titan is hosted by gemini://alexschroeder.ch if I am not mistaken, someone correct me. You should also talk to @skyjake on gemini://station.martinrue.com who maintains Lagrange, the only browser I know of that even implemented titan. Right now, as it stands titan is merely more than an ftp alternative. One that you can add version management on, as if you were running a wiki. I'm not going to discuss the pros and cons of this spec, because I have no need for it. It works against decentralization. And just for my own gemlog, my toolchain would allow me to upload a simple textfile via scp/sftp and everything happens automagically. What I will say however is, try to live within geminispace for a few months, hang around on irc and on station. You'll notice that there's a certain minimalism vibe to everything. People have really great ideas that change your perspective on what you think is necessary and what isn't. On Tue, Aug 31, 2021 at 2:15 AM Sergei Gnezdov 'sergei.gnezdov at gmail.com' <gemproj+sergei.gnezdov=gmail.com@suckless.anonaddy.com> wrote: > This email was sent to gemproj@suckless.anonaddy.com from > sergei.gnezdov@gmail.com and has been forwarded by AnonAddy > <https://anonaddy.com> > Click here > <https://app.anonaddy.com/deactivate/84c97fb7-dd72-4688-b653-188215324c2a ?signature=e4cd6fabb5144657ed72ac3c56cde019ecccfe366af0b08968536f9080c11eba> > to deactivate this alias > > Here is my proposal. It will be in standalone Github project. I am > interested to see the reaction. The point of it is: Gemini as of today can > stay the same and problem of form submission can be done on top of Titan. > > On 8/30/21 4:30 AM, gemproj@suckless.anonaddy.com wrote: > > On Mon, Aug 30, 2021 at 7:43 AM Sergei Gnezdov 'sergei.gnezdov at > gmail.com' <gemproj+sergei.gnezdov=gmail.com@suckless.anonaddy.com> wrote: > >> >> Gemini as of today will not scale in interactive usability, because user >> interaction is severely limited. There is a statement that Gemini is >> only for consumption. However, we have to be realistic that for >> usability of consumption to stay high, user cannot be asked to live >> Gemini browser for HTTPS powered web browser to submit a post. That >> breaks user flow and makes Gemini not worth it. Why to come back to >> Gemini if I just used standard browser to do something and in the >> process I compromised all the advantages listed above? >> > > I know this is not very progressive to say it, but why does it need to > scale? Why does everything always have to appeal to everyone? > Not to mention the well-discussed issue as to whether content *needs* to > be updated through a browser. My stance on this is pretty much: you have to > learn things to use things. And using sftp to upload a file as a bare > minimum requirement is not hard. And it uses a protocol made for file > transfer! You can even transport images with it! </sarcasm> > > In all seriousness, there are multiple real life examples of this. If you > want to learn to drive a car you need to consider learning how to drive. > Who'd think that just because you know how to ride a bike, someone tailors > your car to work like a bike, you know? > > geminispace provides a nice retreat from the complexities of the clearweb > and it also makes you get back to tools that were around before everyone > decided that literally everything needs to live in a browser. It doesn't > mean that it's uncomfortable to use, but as skyjake once put it on > gemini://station.martinrue.com: there's a reason text editors are there, > because text editing is hard. Adding that to a browser is tough. If you > make titan:// mandatory in geminispace, most browsers will instead to turn > to the shell where you have immediate access to nano, vim and emacs, and > what have you gained then? > > I know I am ranting. But it's been discussed. geminispace is feature > complete from a protocol end. And if it doesn't grow big, that's probably > ok. Because the people that are in geminispace are great as it is :) > > >> >> Titan protocol is a good step in the process and perhaps is the only >> required step. I am not sure. We have to remember about need to >> support multiple screen types from desktop to mobile and it is tough to >> create something acceptable and generic for user input under the >> restrictions Gemini protocol put in place. I am hoping people here come >> up with something :) >> >> I also would like to raise people's awareness, that if the network to >> grow and get any close to big, people shall expect that network would >> promote payments models, because it kills advertisements so >> efficiently. I think it is a good thing. I want to consume a product >> and not to be a product. If I am a product, then UI tends to focus on >> makings adds prominent and waiting my time. That goes back to value of >> Gemini and consequence of its proposition is non-add based models. I >> hope. >> >> >> >>
Probably been mentioned before but what is the closest we have to an official it channel? Aug 31, 2021 5:56:01 AM gemproj@suckless.anonaddy.com: > That's dedication! I presume that was a lot of work. > Have you talked to the person who invented titan:// ? I think gemini://transjovian.org/titan[http://transjovian.org/titan] is hosted by gemini://alexschroeder.ch[http://alexschroeder.ch] if I am not mistaken, someone correct me. > > You should also talk to @skyjake on gemini://station.martinrue.com[http://station.martinrue.com] who maintains Lagrange, the only browser I know of that even implemented titan. > > Right now, as it stands titan is merely more than an ftp alternative. One that you can add version management on, as if you were running a wiki. > I'm not going to discuss the pros and cons of this spec, because I have no need for it. It works against decentralization. And just for my own gemlog, my toolchain would allow me to upload a simple textfile via scp/sftp and everything happens automagically. > > What I will say however is, try to live within geminispace for a few months, hang around on irc and on station. You'll notice that there's a certain minimalism vibe to everything. People have really great ideas that change your perspective on what you think is necessary and what isn't. > > On Tue, Aug 31, 2021 at 2:15 AM Sergei Gnezdov 'sergei.gnezdov at gmail.com[http://gmail.com]' <gemproj+sergei.gnezdov=gmail.com@suckless.anonaddy.com> wrote: >> This email was sent to gemproj@suckless.anonaddy.com from sergei.gnezdov@gmail.com and has been forwarded by AnonAddy[https://anonaddy.com] >> Click here[https://app.anonaddy.com/deactivate/84c97fb7-dd72-4688-b653-1 88215324c2a?signature=e4cd6fabb5144657ed72ac3c56cde019ecccfe366af0b08968536 f9080c11eba] to deactivate this alias >> >> >> Here is my proposal. It will be in standalone Github project. I am interested to see the reaction. The point of it is: Gemini as of today can stay the same and problem of form submission can be done on top of Titan. >> >> On 8/30/21 4:30 AM, gemproj@suckless.anonaddy.com wrote: >> >> On Mon, Aug 30, 2021 at 7:43 AM Sergei Gnezdov 'sergei.gnezdov at gmail.com[http://gmail.com]' <gemproj+sergei.gnezdov=gmail.com@suckless.anonaddy.com> wrote: >>> >>> Gemini as of today will not scale in interactive usability, because user >>> interaction is severely limited. There is a statement that Gemini is >>> only for consumption. However, we have to be realistic that for >>> usability of consumption to stay high, user cannot be asked to live >>> Gemini browser for HTTPS powered web browser to submit a post. That >>> breaks user flow and makes Gemini not worth it. Why to come back to >>> Gemini if I just used standard browser to do something and in the >>> process I compromised all the advantages listed above? >> >> I know this is not very progressive to say it, but why does it need to scale? Why does everything always have to appeal to everyone? >> Not to mention the well-discussed issue as to whether content *needs* to be updated through a browser. My stance on this is pretty much: you have to learn things to use things. And using sftp to upload a file as a bare minimum requirement is not hard. And it uses a protocol made for file transfer! You can even transport images with it! </sarcasm> >> >> In all seriousness, there are multiple real life examples of this. If you want to learn to drive a car you need to consider learning how to drive. Who'd think that just because you know how to ride a bike, someone tailors your car to work like a bike, you know? >> >> geminispace provides a nice retreat from the complexities of the clearweb and it also makes you get back to tools that were around before everyone decided that literally everything needs to live in a browser. It doesn't mean that it's uncomfortable to use, but as skyjake once put it on gemini://station.martinrue.com[http://station.martinrue.com]: there's a reason text editors are there, because text editing is hard. Adding that to a browser is tough. If you make titan:// mandatory in geminispace, most browsers will instead to turn to the shell where you have immediate access to nano, vim and emacs, and what have you gained then? >> >> I know I am ranting. But it's been discussed. geminispace is feature complete from a protocol end. And if it doesn't grow big, that's probably ok. Because the people that are in geminispace are great as it is :) >> >>> >>> Titan protocol is a good step in the process and perhaps is the only >>> required step. I am not sure. We have to remember about need to >>> support multiple screen types from desktop to mobile and it is tough to >>> create something acceptable and generic for user input under the >>> restrictions Gemini protocol put in place. I am hoping people here come >>> up with something :) >>> >>> I also would like to raise people's awareness, that if the network to >>> grow and get any close to big, people shall expect that network would >>> promote payments models, because it kills advertisements so >>> efficiently. I think it is a good thing. I want to consume a product >>> and not to be a product. If I am a product, then UI tends to focus on >>> makings adds prominent and waiting my time. That goes back to value of >>> Gemini and consequence of its proposition is non-add based models. I hope. >>> >>> >>> >>
---
Previous Thread: how to make an input field on a capsule page ?
Next Thread: [off topic] For the people still using old 16bit vintage PCS - ngIRCd - IRC Server