New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
The future of the Ruby Toolbox #1
Comments
I like your project a lot. Talking to a community is a great way to start! |
Thanks for all your hard work in maintaining RubyToolbox for all these years. It has been my go to app to check out library options. |
Thanks for opening this thread and for all your efforts to keep the toolbox running! I think it’s a good idea to find corporations who would support a future version of the toolbox. Please leave a thumbs up if you agree and then maybe we can create a new issue to discuss this topic. |
You could also try out scaleway hosting https://www.scaleway.com/pricing/ |
👋 Thanks for opening up this discussion and for all the work over the years! :) 💚 🌟 🙏 Ruby toolbox has been great when it was working so I'd love to see a comeback. Cost-wise, I'm not sure if this is applicable but mabe ruby together would be interested in incorporating the project? I'd consider it part of the ruby ecosystem, I mean it's no ruby gems... but it's a nice accompanying guidebook :) |
I want to thank you as well. The RubyToolbox is a well known project and for me it was always a great place to check for alternative gems. I like PragTob's idea with ruby together - I think you should just talk to them. |
After all these years I still keep checking the Ruby Toolbox to check which gem is the 'hottest shit' right now. And I have the impression others must do too, as I heard many people mentioning the site over time and in recent months. Some even wished to have a Toolbox like site for other languages. As it is seen as such a great resource, I surely hope you/we will find a sustainable way forward. If that will be a complete contributor driven re-write without all time-consuming community features, a cheaper hosting solution, a Patreon financed rewrite or an organisation that would be happy to sponsor... who knows. When you decide to go the latter way, I would definitely contact all larger companies using Ruby that also sponsor conferences and ask directly for a 4 figure some per month for the next year for them to see their logo on the page. I am happy that you are finally going into the direction of open source. I remember us talking about it many years ago. And you basically mentioning the same reasons that have hold your back from publishing the code base. |
Ruby Toolbox is a great resource, but I understand all your stated issues with running a free service like this. I'd be happy to pay a small subscription, but given the experience of Ruby Together who have the profile of the official org for both Ruby Gems and Bundler and yet to hit $20k/mth, I'm probably in the minority. Thankyou for this resource and I support whatever decision you decide to make going forward. |
I haven't used Ruby Toolbox in a few years, but it was instrumental when I was learning Ruby/Rails to figure out what gems were worth looking into. More recently in my career, I started learning Ember.js, which has Ember Observer which I immediately recognized as a modernized RubyToolbox for the Ember community. One of my favorite aspects of it is its review/evaluation system of each package, which asks the following questions (taken from their website):
This is probably more helpful to the still burgeoning ember community than the relatively stable Ruby community, but I dig the idea of identifying which gems have real effort going into them and which are just slapped together to get something done real quick. |
I've found Ruby Toolbox to be extremely useful as someone who hasn't paid too much attention to the ruby community for years and then coming back. For a variety of use cases, like you mentioned, the community has solidified on one particular choice. But how do I know which one that is? This is where I see Ruby Toolbox fitting in, for my use case. I'd love to see that particular functionality preserved in whatever incarnation comes from this. |
First of all, I wanted to thank you for maintaining Ruby Toolbox I appreciate the effort to keep it alive after all these years. Regarding the future, I think what we care most is the ability to compare statistics between different gems with the same or similar functionality. One of the options that I will like to see in the future is a distributable command line tool; I don't know how you collected the GitHub repo info, and the score but that could be done inside the distributable file. |
I think a modern Ruby Toolbox still has a role and I'd be happy to make a small monthly donation towards upkeep and hosting. Hearing that you were paying 60 EUR per month to host it alarms me. I've had an opportunity to learn more about technologies like AWS and Docker over the past year, and I do know of ways to get a web app hosted for much less, even with the app having multiple backing services. I'd be happy to share those techniques with you if you like. |
The Ruby Toolbox was a great resource and I would find it useful today if it was still around. The core concept was solid, the only things I'd suggest would be an easier way to add new gems and maybe an API. If you started up an OS version I'd gladly contribute. Thanks for all the work you've put into it! |
|
👋 btw. I reached out to Andre Arko from rubytogether and he said he'd bring it up in the next board meeting :) |
Personally I don't belief that rubytogether would be a viable option to get the Ruby Toolbox back quickly. AFAIK they bill 150 US$ per hour and don't do free work under their brand. As they get less than 20k per month, they just have around 120h/month to distribute for all their projects. But let's see what they conclude in their own evaluation. |
What I would do in your case is: release the source code as it is, find someone I would trust to deal with the community and maintain the project - Ideally someone who would benefit from the extra visibility but is still somewhat experienced. I would then assume the role of a benevolent dictator, only stepping in when there are conflicts / excessive bikeshedding / abuse (and maybe creating some basic issues for things that need updating to get the ball rolling). To be clear, I am not volunteering to do the community stuff. I've got a 11-month son :) |
@dwradcliffe do you think shopify could be interested in something like this? it sounds like a great hackdays project to me. |
I think something simple with categories, scores (based on some metric) plus some Git/repo stats would go a long way. Simple search would also be useful. Basically just a rebirth of the Ruby Toolbox (I've used it many times, but realise it may be time for a refresh). Perhaps inspiration can be taken from either of these two: Development Are you open to using a client-side framework for portions of the site, or do you want to stick with server-generated views? If a client-side framework is okay, I may be able to help with something built on Ember. It's quite easy to hook up to an Rails-API. As for hosting, if it's built as a single-page app with data in S3 it'll be fairly cheap. For search one can use Algolia (https://www.algolia.com/pricing), they have a community/free plan for open-source projects. |
I have used The Toolbox for years and it was always helpful. I am happy to contribute to a fund if that helps |
Thank you for all your hard work over the years and I hope you will consider moving forward. RT has been a great help over the years helping me switch over to Ruby/Rails from other languages. Count me in to help out if you go open source. As you can see there is still a great amount of interest in the toolbox. Maybe do a kickstarted campaign so you can devote some hours to getting it back up on its feet 100%. Whatever you choose to do thanks again! |
Thanks for your work on this, Christoph ❤️ makandra would be happy to sponsor servers for Ruby Toolbox, either in its current form or whatever it becomes in the future. It would be a managed setup with redundancy, so you wouldn't have to worry about Linux updates or broken hard disks anymore. Our ops could also help you with the migration if you like. If that becomes interesting at some point, send me an e-mail. |
Wow, thanks everyone for the great, encouraging responses so far! I'll try to spin off some of the topics brought up here into separate issues tomorrow so we can have more of a structured discussion on specific things there. Thanks again for your all your great input, and keep it coming :) |
I just want to add in: I had noticed that Ruby Toolbox was languishing, but I've found always it an incredibly valuable resource. Especially useful when you're coming back to a category of gem and want to know what's the current state of the art. Possibly even taking it beyond Ruby into JS and other languages would be great too. |
I've taken a lot of inspiration from Ruby toolbox with https://libraries.io, which tries to help solve open source discovery for every community, one area that I've not had chance to get up to speed on yet is categorisation. We're a non-profit project funded by the Sloan and Ford foundations and if there's anyway I can help do let me know, either by providing data to refill rubytoolbox or adding the same categories into libraries.io |
Thanks @andrew, your work on libraries.io is very impressive! I think focusing on just this one programming language was actually providing a lot of value to users of the Toolbox. I guess it depends on the use case, if I just want an awesome postgres CLI tool, then I don't care what it's built in. When I need authentication for my rails app, going with Python might be cumbersome. Even supporting this one language as a maintainer proved to be very difficuilt in the long run, it's just a lot of effort to maintain the categories in a sane way, keep up with new stuff etc, and even display stats that are most relevant in that exact domain. Introducing and maintaining categorization (and a schema for moderating them) would be much harder for 30 package managers than just one programming language. Two things the Toolbox (imho) did well in it's heyday was discoverability, through both the search and the categories, and usability - I always tried to show the most relevant stats front and center wherever you saw the project, so there was no need to click through to projects from overview lists. Take for example the following case: I want to find a ruby code coverage library. On the Ruby Toolbox search, this showed up the most popular ruby gem simplecov first. When I search on libraries.io, I get:
Don't get me wrong though, your site is awesome, the only thing I'm getting at is that I think there is a difference between a general-purpose open source database, collecting allthedata(tm) and a site that tries to help aid in finding what's used by everyone else for any given problem. Anyway, the data that you are syncing is definetely also what the Toolbox is using. @andrew Could you maybe give more details on the "Open data release coming soon." that is announced on the front page? It might indeed make a lot of sense to use your db as the backbone? On the other hand, I'm a bit skeptical about using a "proxy" API instead of the proper upstream APIs, since there would be an additional moving part that could break, and keeping in sync might be tough (consider rubygem release webhooks for example), but that all depends on what integrations you have planned :) |
Our open data release is now public, blogpost here: https://medium.com/@BenJam/libraries-io-releases-data-on-over-25m-software-repositories-ab1db665826e, more info and documentation here: https://libraries.io/data and the actual csv downloads are here: https://zenodo.org/record/808273 We also have an API on the site here: https://libraries.io/api, if there's anything missing we can definitely add it, we've also got a basic streaming API where you can listen for new versions, which will be beefed up with all repository events over the coming months. All the data is available under a CC-BY-SA license and all the code for Libraries.io is open source too. Agreed, our search isn't great at the moment, we've been focusing on the goals of our grants which mostly involve collecting more and better quality data (full list of grant goals here: https://github.com/librariesio/supporters/issues), but discovery is very important to me, it's why I first launched the site. Another area that I see happening a lot with open source is a developer builds a tool for one community (usually centered around a package manager) but it's too closely tied to the implementation and can't be reused by any other communities, so there ends up being a lot of rework and a lot of dead projects. I've been trying to normalize the data for all the different package managers and source code repositories to help people build tools on top of the data that work for everyone, as well as avoiding a GitHub based mono-culture, which is why I've spent a lot of time adding support for collecting metrics from GitLab and Bitbucket repositories as well. When it comes to the categorization, it's potentially quite a fun problem to solve to try and get communities to categorize new and updated projects as they are released, potentially with some automated classification based on keywords, etc Happy to work with you in whichever way you think is best! |
@andrew That sounds awesome, I know by heart how hard aggregating this data is, really great work! This might be indeed extremely helpful, in whatever way we'll proceed. I'll get back to you, but thanks a lot already for your input! |
I don't have much to add to the discussion, though I'll follow it with interest and would consider time or money in support. I just wanted to say ruby toolbox remains basically my first stop when I dig into a new area of ruby or rails gems - it might not always by itself be sufficient info these days but it's still by far the best place to get a quick lay of the land. |
Hi all, so many people want this to be a thing. 🙌 Can we please make steps? If not i'd consider closing this issue. Sorry for being so rash sorry not sorry, but i think it is proven that people want/need something like this. So building it is paramount, and i don't think we really need to talk a lot more about it, we need to build it. I was thinking of maybe organizing a big hackathon, and then just hack it over the course of a weekend. Or at least launch something? I don't mind helping organizing it. I'm not saying it will be done then but a start is made. If the hosting costs of 60 a month is an issue. I can chip in 30 and i'm pretty sure my founders or my fellow devs are also willing to pay. So i don't think thats the issue, lets just build the thing and stop bikeshedding. Thanks a lot <3 |
I do not think building |
Why not? I believe the aggregators and counters werent up to date and a lot of it was stale anyway. Not launching and building it because the initial data is missing seems like a missed opportunity. I dont think its impossible to build it up again.
Nowadays with github dependents and gem info we can sort of rebuild it again clean slate
…Sent from my iPhone
On 14 Aug 2017, at 15:43, Chris Hough ***@***.***> wrote:
I do not think building new is the issue, rather the old statistically data that made the toolbox what it was. A clone without it, would not be very valuable at all.
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub, or mute the thread.
|
Nowadays with github dependents and gem info we can sort of rebuild it
again clean slate
And that is what we'll do. Real soon now.
|
Access the ruby-toolbox site from 16th May 2017 here https://web.archive.org/web/20170430093635/https://www.ruby-toolbox.com/ |
@colszowka Thanks for creating this site that the Ruby community (myself included) has appreciated for years! Based off the publicly available traffic stats for Ruby Toolbox, you should be able to get $250 - $600/mo using AdSense alone. I suspect there would be other opportunities for additional revenue as well given your position in reaching out to developers. A couple banner ads on a page is not going to degrade the site. In fact, I see it as the opposite -- the revenue justifies and incentivizes the time necessary to make the site even better. |
@colszowka -- We use it! We need it! We love it! You're awesome! |
I love the site. I would love to be part of updating the code and even donating. I wasn't aware that the costs of running it was that high. I don't think that a tool like this will benefit from adds, so I will happily be a Patreon for a few bucks a month. And I suspect -- based on the level of love in this thread -- that others will to. Perhaps if we start an open source version, we can also work on getting the price down by selecting better an cheeper tools as a community? |
Out of interest, how much server resources do you need? In terms of number of CPU cores, RAM, disk space? I'm the CTO of the company behind https://www.civo.com and while we haven't talked internally about sponsoring OSS projects with free hosting, I'd be happy to have a conversation with you first about whether it's feasible... Feel free to drop me a mail - andy@absolutedevops.io |
Just found this meta-ripoff of the toolbox that works for many languages. |
One of the really valuable things that Ruby Toolbox had was the at-a-glance information about
That is what constantly helped me wade through lists of gems that all claimed they could be helpful. That is what distinguished Ruby-Toolbox from other lists of gems ('awesome' or otherwise). My input is that this is a key feature that must be included. Perhaps this would be something that we could start on now? Some folks could put together some different UI possibilities, and some can work on the 'getting the data' part. Maybe folks can start with the existing code, or maybe there's value is also starting from scratch..... ? |
Hello again everyone, time for an update! As you might have already heard on twitter last thursday, @rubytogether have offered to sponsor my work on re-building the Toolbox and will be funding around 5 hours of my work on that per week - awesome! I was in Budapest for EuRuKo over the weekend but will be getting started on this now. The general roadmap for the next weeks looks like this: I want to bring back the core functionality of the site back online as soon as possible, so we have a good starting point for future features and improvements. One thing I have realized over the past months is that it's probably near-impossible to build up a community of contributors without an existing site and code base to talk about, since issues are not tangible enough and too broad. So my focus for the next weeks will be to bring back a baseline Toolbox that anyone interested can start contributing to. I wrote above and also for example in #4 what I consider the core features, and from I think this is supported by most comments made in this regard in this issue:
I've decided to, at least for now, go with plain rubygems.org and Github APIs as the data backends since after a deeper look into the libraries.io APIs and available data (see #5) many metrics I consider important are not available in their data set. I still like the idea of acquiring data from libraries.io since this would enable other language communities to build on this code base, too, but like I said I also want to bring back the Toolbox with a decent feature set as soon as possible, and I think only the direct access to rubygems and github can provide that at the moment. Swapping data providers down the road should not be too hard though, so once we're in gentler waters we probably should re-evaluate and sync up with the libraries.io team. Regarding data, I'll try to go for the low-hanging fruit first, meaning the data that is available directly in the upstream APIs. Regarding the categorization of projects, the old Toolbox used to have a built-in feature that allowed to send in change requests for projects, i.e. adding a category. To avoid the burden of maintaining this feature in the future and to allow easier community contributions, discussion and reviews I set up a new repository at https://github.com/rubytoolbox/catalog which contains the Toolbox's categories in a straight forward YAML file and directory structure. The repo holds a README containing a description of the format. I'll be working on adding a CI parser for validation and export of this soon. If you'd like to help, this is where you can get started right away! As I've said before, the categories have seen too little maintenance love from me in recent years, so we have a lot of spring cleaning, adding of new projects (and removal of dead one's?) to do, and this can begin right away over at https://github.com/rubytoolbox/catalog This got longer than expected once again :) I'll try to get better at giving more incremental, short updates, but I felt it neccessary to give everyone an overview of the current status. Finally, thanks again to everyone for your encouragement, your kind messages, 👍s and ❤️s mean a lot to me! Cheers, |
Very good news! I'll try to incorporate the data from my lists into the Catalog @colszowka please look at this issue: #8 |
This is great news |
Yay to sponsorship news! Just wanted to point out that OpenCollective.com might work for you as well -- it's a site that helps unincorporated open source projects handle the tax implications for open source work of certain types -- OpenCollective Foundation is a 503(c)3 that they run to be a "fiscal sponsor" (ie. loan a slice of their bank account) to open source projects. Might be a way to slowly spin up sustainable funding in recurring donations, for if/when @rubytogether support runs out, or in case you need to bill more hours. @piamancini can probably clarify if I got any of the facts out of whack :) |
@rubytogether is a host on open collective so they could hots this collective. Otherwise, the Open Source Collective would be happy to. |
For the record, @rubytogether plans to fund this work in a sustainable and ongoing way. The more developers and companies chip in by becoming members of Ruby Together, the more work on Bundler, RubyGems, and Ruby Toolbox we will be able to fund. And if @colszowka wants to run an Open Collective just for Ruby Toolbox, we would also be happy to host that collective at Ruby Together. 😄 |
I'd join an open source project. I need all the experience I can get. |
@merlin2049er Head to https://github.com/rubytoolbox/, fork a project, check the issue tracker, comment, code, push, repeat. |
Hi, Platform.sh will be happy to provide free hosting for this; Just sayin. If interested contact me. O. |
Hey everyone, the site is back 🎉 😀 - https://www.ruby-toolbox.com/blog/2018-02-01/lets-push-things-forward Best, |
Oh man, the new site looks very sexy… |
Awesome work! Thank you! |
This is awesome news! Super stoked to see this. |
Thank you sir !
… On Feb 2, 2018, at 04:23, Chris Hough ***@***.***> wrote:
This is awesome news! Super stoked to see this.
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub <#1 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AApC2v2BkfDkQNPSVLL90yE903ddqkBpks5tQjz8gaJpZM4N45Qe>.
|
Hello everyone,
as you might have heard, last week the server hosting the Ruby Toolbox crashed. While it's definetely possible to re-build from backups, I would like to use this opportunity to set in motion a discussion about the future of the site.
Launched in 2009, the Toolbox has always been a closed-source site. There's a bunch of reasons for that, but the most notable one was that I did not want to take on the burden of having to manage an open source community around it, having seen how much of a time-sink maintaining SimpleCov has been for a long time. This view has changed over the past few years as a bunch of smart and handsome people joined the simplecov team as contributors. I also believe the general attitude of open source users has shifted, caused by a whole bunch of "open-source-burnout" blog posts to a more positive, and encouraging community, replacing a wide-spread feeling of entitlement to free support.
In 2014, my wonderful son Noah was born and as surely many parents can confirm, it becomes somewhat harder to hack away into the night, so even though I've had plans to open source the Toolbox for years, I never managed to get this going.
Over the years, I also got contacted by literally dozens of people asking to help, but since I never reached a baseline, published open source repository we could have built upon together, I never could really give a pointer where to start.
Why not simply open source the current version?
Why is it so hard to build a new open source version?
It's definetely not hard to build a fresh version of the Ruby Toolbox, from a technical standpoint.
More than anything though, every time I attempted to build a new version of the Toolbox, I stumbled into "philosophical" issues. Is this even needed? Who would use this nowadays?
What should the Ruby Toolbox be?
Since I started the Toolbox in 2009, the Ruby ecosystem has changed, and stabilized, a lot. For many use cases, the community has settled on a specific solution. I would really like to hear what you all would like too see from the Ruby Toolbox. I won't go into much detail on my ideas in that regard here in order to not give too much direction into this discussion.
Sustainability
I've experimented a bit with adding ads to the Toolbox in early years, but the 100 - 200 USD per month were in my opionion ultimately not worth polluting the site with banner ads. For a few years now, the site has had a Paypal donation button, which currently generates ~25 USD per month from 3 recurring donors and one-off donations between 0 and 15 USD per month. I've been taking care of paying the hosting, which has been around 60 EUR per month.
I take great pride of providing a resource that has proven very useful to fellow developers all over the world, but it also is sometimes down-putting to, beyond time for development and maintenance, also add money to keep the site going.
After the outage, plenty of people suggested to create a more formalized donation scheme, for example with Patreon. Would people be interested in this? Ultimately, if we get enough donors I even could put in some proper work time in place of my usual freelance consulting gigs. I also wonder whether companies would be interested in joining in on this? Again, I'd love to hear your thoughts!
I'm looking forward to your feedback, and thanks a lot for the many kind words of encouragement and support that reached me over the past week! Let's get this started!
The text was updated successfully, but these errors were encountered: