search

Shopify / slate

Slate is a toolkit for developing Shopify themes. It's designed to assist your workflow and speed up the process of developing, testing, and deploying themes.
https://shopify.github.io/slate
MIT License
1.28k stars 364 forks source link

update on project priority and plans for next steps #1063

Open ild0tt0re opened 4 years ago

ild0tt0re commented 4 years ago

Dear Shopify Team, from your last announcement

This low maintenance period is time-boxed to 6 months from March 2019. We will re-visit and re-evaluate the project priority and plans for next steps.

More than six months have passed and we would like to hear about this project as we are evaluating whether to use Shopify platform for our customers.

We consider this project of primary importance. Thanks.

javy- commented 4 years ago

@t-kelly is the version bump indicative of development resuming?

https://github.com/Shopify/slate/commit/d3b4cd867f91d0c8b56c61f438c096090dee1c59

t-kelly commented 4 years ago

No -- that was only me jumping in for some quick wins. Still waiting on the official word of how things will proceed.

lmartins commented 4 years ago

I think it needs to be said, that the way Shopify is treating the developers who invest and work with this platform is wildly disrespectful here. I'm sure there are bigger reasons here, and perhaps something good is coming down the road, but no way this is a proper response to people who help build the platform.

joeldmyers commented 4 years ago

I agree, and I think Shopify does need to consider whether it's worth investing in its development community. Developers are often the ones deciding which platform and technologies their clients should choose, and to be totally honest Shopify doesn't have a great reputation among developers right now. Part of that is ignorance as to how robust the platform can be. But part of that is the direct result of this kind of neglect, and not keeping apace with state of the art development. It definitely does not reflect well on what's an approx. $34bn company, that their primary way of making custom themes seems to have been abandoned, and Shopify will definitely lose a lot of money in the long run from developers who decide to choose other platforms for their clients, if there isn't adequate support and tooling available.

cgilbe27 commented 4 years ago

You don't necessarily have to use slate; to use npm packages, CI/CD, E2E, asset minification, visual testing, or linting. We have accomplished all of it. Slate mainly provides direct tooling with Shopify along with some built in linting support and deploying, nothing unique by no means. In its current state, Slate is not usable for full development, you can use it, but don't rely on it. Also, the secure localhost issue was a huge problem we ran into, which was the main reason we went with a piecemeal solution, it also resulted in a better comprehensive solution, and dropped Slate completely.

joeldmyers commented 4 years ago

Ah yeah makes sense; I've put in a lot of additional things (linting, E2E / unit tests etc.) but kept Slate for deployment... I might just need to do the same and build something out for deployment on my own. Thanks!

cgilbe27 commented 4 years ago

I strongly recommend deployhq, awesome deploy service and great technical support

lmartins commented 4 years ago

@cgilbe27 while I'm curious as to how you use deployhq to achieve an equivalent workflow allowed by slate (livereload, deployments, etc.) lets perhaps keep this issue on topic and that is, the lack of information coming from Shopify on the future of this project.

curiouscrusher commented 4 years ago

While it's clear we all want to have some insight on what the future of Shopify's developer offerings will be (Both with Slate and other toolsets), it's important to keep a few things in mind.

First, Shopify's engineering staff more than likely does not get to set the priorities for company wide initiatives. So while venting into a GitHub issue on an open source project might make you feel better, it's highly unlikely to truly accomplish anything useful at the company level. Secondly, this is an open source project, it's provided as-is with the intention of helping developers who may not have time/resources to hand-roll their own solution. It's also a great learning tool that can give insight into how the Shopify team builds and architects tools/features.

If you're unhappy with Slate, there's a very large list of alternatives out there. And if none of those meet your needs then I suggest you build your own development workflow and tooling, that's what plenty of others have done.

cgilbe27 commented 4 years ago

@lmartins Basically my point was that while Slate has features, they are definitely not hashed out completely, (I mean just look at the amount of issue requests). So from an objective point of view, if you want features outside of Slate or something in Slate that, well, isnt broken, you'll need to tie those in yourself, but that its very simple to do so. OP stated they were weighing whether or not to go with Shopify, but I dont think Slate should be a reason to not go with Shopify for your site. Also, themekit and prepros does great for JS and SCSS minification + it includes browser reload and is suggested by Shopify. An update would be nice however, because we would likely want to extend Slate in the future once its at a non-beta stage.

scottsb commented 4 years ago

@curiouscrusher Venting here may not accomplish much, but this issue has visibility up to the CEO:

June: https://twitter.com/tobi/status/1138581134156161025 November: https://twitter.com/tobi/status/1191715755127136256

drinkmorewaters commented 4 years ago

@t-kelly Just to say, that the only reason i can take any enjoyment developing for Shopify is Slate. Even its not perfect, having a local development environment, is just much better than using the Shopify editor. We use Vue a lot and its working OK. Any modern development pipeline is appreciated and the work put in doesn't go without appreciation.

There's a redesign of the theme engine going on, bunch of changes to sections and the likes, lots of API's changing. I think people just need to wait until Shopify has finished those before Slate can get some redesign too.

It would be great if the front end development moved to something that was able to be moved outside of hosting within Shopify, without having to use the graphql API, a solid front end theme SDK. Pick Netlify, Zeit, what have you, build how you want. Coupling the liquid with the front end has obviously made things very hard to move quickly for Shopify.

Thank you for the hard work!

amosmos commented 4 years ago

Hey all and especially @t-kelly,

Just wanted to say how great Slate is already now, and how great it was to transfer the SodaStream set of Shopify stores to Slate.

We launched it today and it's already showing amazing performance results!

Thanks!!!

chanmathew commented 4 years ago

What's everyone using now that Slate is in low maintenance? I've only found Slater so far....are there other alternatives I'm not aware of?

curiouscrusher commented 4 years ago

@chanmathew I’ve explored slater as well as some other options like Motifmate and even some of the Plus Partners house rolled solutions. I always end up back at Slate because it’s just so well rounded and fits all the needs of flexible Shopify theme development. Even on low maintenance it’s very much a complete solution, and since it’s open source you can customize the code to suit your needs. Most of the other solutions are too targeted or opinionated for my needs and I’ll end up spending more time customizing them than I would have spent just starting with Slate.

aeu commented 4 years ago

It is now January 2020, nine months since the 6 month time-boxed period. Is there any update available regarding the status of this project?

javy- commented 4 years ago

Slate - End of Support (January 2020)

For anyone following this thread who didn't see the update yesterday...

drinkmorewaters commented 4 years ago

Thank goodness Shopify finally had the (Apple courage) to come out and put Slate down. To @t-kelly your hard work was everything. Appreciate the effort.

Google EOL meet Shopify EOL. Two pees in a pod.

lmartins commented 4 years ago

Agreed. Shame how long it took for them to come clear on this, but glad they finally did it. I don't think at this stage these news have caught anyone by surprise.

adrianocr commented 4 years ago

Ok so you guys say that Slate doesn't solve the problem of local development and theme versioning. Cool, I agree. But is that that? Or do you guys have any plans to do something about it? Because your statement basically comes off like "Yeah it doesn't work how we'd like so we're ending support. Good luck." and that's it.

Are you guys happy enough with the offerings from Maestroo and Pixel Union that you think they fit every customer's needs? I think it's ill-advised to alienate your developers like this. There's what, a few hundred of us? Few thousand? Are we not enough of a financial incentive at Shopify that you guys feel you can ignore us?

I've personally built 52 Shopify stores to date and about ~48 of those were custom themes. 12 of them Shopify Plus stores. Many times we end up pushing the customer towards Shopify because we like the platform. But if Slate is no longer supported then building themes on Shopify becomes archaic again and we have little to no incentive to not go with WooCommerce where yeah, WordPress has its pitfalls but at least building themes for it is a modern and easy experience. Bigcommerce has its issues but at least building themes for it is a modern experience.

Yes I'm being hyperbolic here but what I'm trying to get at is: how about some better communication? How about a ballpark road map? Hell, I'd even take a ballpark for an announcement date for a road map at this point.

lmartins commented 4 years ago

@adrianocr I've been where you at. Super frustrated with how they have treated developers here. I'm done with that.

They are saying that you can use themekit and build on top of that. Personally I'm using theme kit and my own Webpack config to achieve something that fits my needs. Maybe this whole situation with spark the appearance of new tooling, and maybe Shopify will have something new in the future. Either way, the way they have conducted this was very poorly managed.

curiouscrusher commented 4 years ago

Well that sucks, far more drawn out than it should have been. And I can’t say either of the stated reasons for sundown even seem related to Slate all that much.

Seems like the vision is far overreaching the actual need and problems that ahould be solved. A simple, reliable, and maintained theme development tool would likely go much further with the community than some grand plan to unify local development, new sections, and theme versioning. As much as I’d love to see proper version control for Shopify Themes, we really just need a steady tool to work with for longer than 6-12 months. Especially if Shopify expects the development community to grow, or even stay steady.

drinkmorewaters commented 4 years ago

@adrianocr I've been where you at. Super frustrated with how they have treated developers here. I'm done with that.

They are saying that you can use themekit and build on top of that. Personally I'm using theme kit and my own Webpack config to achieve something that fits my needs. Maybe this whole situation with spark the appearance of new tooling, and maybe Shopify will have something new in the future. Either way, the way they have conducted this was very poorly managed.

It would be ill-advised to build your own tool set for stores that have limited management, considering theme-kit, could be EOL at any stage, and the new theme architecture could render whatever tooling you build useless at any point.

lmartins commented 4 years ago

It would be ill-advised to build your own tool set for stores that have limited management, considering theme-kit, could be EOL at any stage, and the new theme architecture could render whatever tooling you build useless at any point.

Yes, Shopify could pull the same move on Themekit and that would suck. But to say that it would be ill-advised to know your tools and explore Webpack to serve your purpose feels a fit of a stretch. You can set it to have hot reloading for JS and SASS/CSS, and you probably can setup Browsersync to do the same for liquid files.

Am I happy with how things unfolded? Absolutely not! I've expressed many times on this forum how disrespectful this whole ordeal was towards developers, and aside from @t-kelly who gracefully tried to give us some context, I haven't seen any good reasoning behind this. Much to the contrary, as @curiouscrusher mentions, this decision only contradicts their arguments to get slate on a low maintenance in the first place.

With all that said, it's the web, you have options, both on tooling and on e-commerce platforms too.

drinkmorewaters commented 4 years ago

It would be ill-advised to build your own tool set for stores that have limited management, considering theme-kit, could be EOL at any stage, and the new theme architecture could render whatever tooling you build useless at any point.

Yes, Shopify could pull the same move on Themekit and that would suck. But to say that it would be ill-advised to know your tools and explore Webpack to serve your purpose feels a fit of a stretch. You can set it to have hot reloading for JS and SASS/CSS, and you probably can setup Browsersync to do the same for liquid files.

Am I happy with how things unfolded? Absolutely not! I've expressed many times on this forum how disrespectful this whole ordeal was towards developers, and aside from @t-kelly who gracefully tried to give us some context, I haven't seen any good reasoning behind this. Much to the contrary, as @curiouscrusher mentions, this decision only contradicts their arguments to get slate on a low maintenance in the first place.

With all that said, it's the web, you have options, both on tooling and on e-commerce platforms too.

What i meant, and it still stands, is that it is ill advised to build tools for clients of which you won't maintain. So building anything for a client where in 6 months, Shopify could kill it by changing their own open source tooling, is and always will be ill-advised. Unless that client can afford maintenance or repairs.

From this day forward, if you use Slate for a client, that's ill advised, if you build something custom for a client that you won't be maintaining that tooling for on top of Shopifys tooling, that's ill advised.

And if you trust Shopify to build standard, consistent, reliable tooling, with great documentation, at the level of something such as VueJS or React, that's ill advised.

lmartins commented 4 years ago

@drinkmorewaters and you're still not getting what I meant but I couldn't care less. Drink more water brother.

drinkmorewaters commented 4 years ago

You have no point, this is about Slate and Shopify killing products that thousands of stores and potentially hundreds of thousands to millions of people might interface with, go learn webpack somewhere else, for commercial use, don't setup clients for projects that get EOL or could break at any time of which, they can't afford to maintain. As in 12 months time, you won't have maintained a thing, making your webpack the slate beta 2.0. Nothing better at all for that client. If building a toolset for Shopify is easy for you, you take over Slate. These actions by companies are atrocious when they don't think that developers, shop owners and their staff paid money and poured in time upon their toolset. If you built a Slate store, its effectively a write off now if you were relying on it's build steps. Unless you have the time to fix or build something better, aka slater.

Shopify needs to take a look at VueJS, React, Ionic, Gatsby, the entire AWS suite, Salesforce. List goes on. Do not EOL a product without an alternative, or at least plan many years out knowing that the investment is worth it, some of these tools Shopify builds might not end up lasting longer than 2-3 years. AWS is so good at this, S3, cloudfront and EC2 are 10+ years old now, still rock solid.

Shopify launched in 2004. Just pick a single tool-set for themes and stay with it already.

But to say that it would be ill-advised to know your tools and explore Webpack to serve your purpose feels a fit of a stretch.

lmartins commented 4 years ago

@drinkmorewaters all I said is that if you do this for a living, you have alternatives and you can kinda make it work until something better comes up. I'm not here to learn Webpack, I know where to look for help on that and I'm as entitled as you to come here for help on Slate.

What I did was offering a different perspective, something that you seem to have a problem with. That tells more about you than it does about me.

Whatever tool you use to upload, the end product is the same, a damn theme you can take from there any way you like. If that doesn't work for you, then yeah, go cry in front of Shopify headquarters or something, because that's what you have now.

adrianocr commented 4 years ago

Maybe you two should go argue in private.

The point stands that at this point Shopify still refuses to communicate with the developers who pour hours upon hours upon hours developing themes and stores for their platform, refuses to give us even a hint at future plans, a road map, and is all around jsut treating us like an unimportant cog in the wheel. Do they think they're too big to fail?

And yeah themekit works but for how long? Next thing we know that starts throwing out errors, issues pile up in that repo, and we get another EOL announcement.