Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

This sounds way too good to be true.

Specially considering the stories from other people here, which include a lot more technical detail.

Also the other users are not new accounts with just one comment (unlike the user above).



Am I more convincing, with a six year old account?

I'm the exact target market for Gatsby, and can confirm the benefits that GP gave. Perhaps I can tell you how I've been selling it to my clients. I work at an agency, which builds sites in the ~$100k range. Most of our clients are currently using something like Drupal, or an ancient proprietary CMS. The RFPs now often specify Drupal "or other open source CMS". We're now offering them Gatsby + headless Drupal, and it gives them some significant benefits. These are sites that are currently paying thousands per month on multiple load-balanced Drupal servers, or managed hosts like Acquila. Gatsby lets them switch to a single Drupal instance on something like Lightsail, firewalled to give access to just their admins. Gatsby can then build the site (using CodeBuild or Amplify) and deploy to S3, so their hosting fees are then tens of dollars in CloudFront egress plus Lightsail. They also don't need 24x7 Drupal support, because it doesn't matter if the server ges down in the middle of the night, because their site stays up, and they don't need instant Drupalgeddon patching because their server isn't public-accessible.

I think the classic JAMStack/Markdown features of Gatsby are a red herring. I love them for my own site, but no client will use them (even with Netlify CMS, unfortunately). However it gets developers like me to try it out, and hopefully realise the benefits that it can offer.


My biggest concern with Gatsby is the build time. Drupal is not exactly known for being fast, and as I understand it, Gatsby asks for the entire site's worth of content for every build, right? How has that worked out for you?

I have Drupal sites with dozens of entity types and tens of thousands of nodes. I feel like Gatsby would have us waiting for an hour-long build every time we want to post a new node.


To be fair, I think the guy above you might be talking about different use cases.

If it's just a static landing page I think the Jamstack/Gatsby/Gridsome would be fine.

If it's anything more complex though, you have my condolences. Ecommerce honestly sucks for the reasons others have mentioned.

I've almost finished a Shopify build with Nuxt.js and there have been so many hard learnings.

The more painful one was storing checkoutIds using SSR and cookies. I still don't know if it's best practice or not :/


I definitely understand the disbelief, but this is the truth. I started visiting hacker news to try to familiarize myself with the tech world, learn new jargon, blah blah blah, and have always just lurked. I think so highly of Gatsby that I finally created an account just to say so.

Like I said, I come from a graphic design background. Mostly freelancing and working for different organizations that just needed an in house guy to do a little bit of everything. Low pay stuff in non-tech organizations - literally a small boutique in one case where I also managed retail workers. I never really got a good career off the ground for many reasons, one of which was probably the "great recession". Not a ton of demand for mediocre graphic designers who aren't very career oriented to begin with.

Probably around 4 years ago I got it in my head that maybe I could learn to build websites and either get a side hustle going or work my way into it. I did online courses, videos, etc., but never had the money or guts to jump into a code camp or something more legit, and honestly nothing really stuck.

Started a new graphic design gig around 3 years ago. The company was in Real Estate. Like I said - large sales team that really needs their own mini-sites within a site as the service we provide is really driven by their personal brands and relationships. As a designer on a small design team I mostly produced print ads for the sales folks - we were like a mini agency and a way of recruiting top talent. A strong corporate brand / web presence hasn't been necessary to create a successful company of just under 1000 employees, although everyone involves knows that it would only help the business. Pretty shortly into my time there I watched the web team (this includes a project manager, front end developer, ui/ux designer, 3 person SEO/analytics team, a content/data entry person, a middle manager for all of them, and some resources from the IT department) fall apart. That group wasn't really producing much - they refused to even build landing pages, didn't do any kind of SEM, etc. This is for a variety of reasons many of them political. Maybe this entire environment seems a little crazy, but I think this is far more common in many industries than people involved in tech might realize.

At some point an agency was brought in to manage the web presence, but they were very obviously neglecting our site. Managers in my department knew I had done some web in the past and used me to run around the agency. The agency kind of knew it, and even had me do stuff on their behalf - I didn't care it beat designing ads for local newsletters, and mailers. A really generous javascript developer at the agency mentioned Gatsby to me once, saying that he wanted to build everything on it, including the next iteration of our website. I could see the writing on the wall, so I decided to learn it thinking(hoping) that I might have to maintain a Gatsby site one day. I think 1.0 had just been released at that point. Not long after, they were fired and I was left holding the ball.

The site I inherited was built on wordpress. Because of the neglect it was many versions out of date, and rotting the way wordpress sites can when they aren't maintained. I did what I could to hold it together and doubled down on learning Gatsby and React. When I started replatforming our website it was a lot of false starts, I played with Hugo, headless wordpress in Gatsby, all kinds of schemes - honestly figuring it out as I went along. Eventually stuff started coming together. Each feature I was forced to build (was making an exact duplicate of the wordpress site) forced me to learn more about the toolset that was organically coming together - gatsby (which includes react and graphql), netlify, contentful, mailgun (remember that complicated form routing/logic), and more. I asked the IT dept, what the wordpress instance for our og site was costing us and learned it was $500+ a month. I did everything I could to keep us in Contentful and Netlify free tiers and have done so. We do pay $15/month for a service that does some form routing/templating magic for us - but that's the only cost these days.

In the end it took me around 6months to do something that would have taken an agency a month to fulfill. But, I also helped on a lot of random projects (it's amazing what people will let you be a part of when you are eager), and served as webmaster for the first site. During that time leadership at my company wasn't really ready to think about web, so in the end I have provided a great value - a huge cost savings for them over what they were spending and even what they otherwise would be today. I think a lot of people here would cringe to look at my code. It looks like what was written by who it was - someone learning as they went. But, nothing ever breaks, site hasn't been down once since replatforming and moving to netlify, site traffic has increased, we finally pass security audits, page load time went from 8seconds average to somewhere around 1-2 seconds on average. I spin up landing pages regularly, and have plans for meeting unique industry needs in a way that is doable and maintainable by just me.

At the end of the day I'm a novice who's making it work on my end mostly by force of will in an industry where a web presence is only just starting to matter. BUT none of it would be doable without Gatsby. Their community, documentation, and product are all amazing. I'm convinced that if my story didn't include Gatsby I wouldn't be provided the opportunities I currently.

If you want to be cynical you can interpret all of that as 'guy at shitty company, in shitty industry, where web really doesn't matter, uses Gatsby to make site that just works, and saves a marginal amount of money that definitely had to be spent (hosting costs) and a large amount of money that probably shouldn't have ever been spent (bloated web team and agency that sat on their asses).'

from not enough detail to waaaaaaayyyyyyyyyy too much


500 -> 15 dollars a month. So not really 0.

I'd also state, that is, for now. These services will have to make money at some point.


You're absolutely right, not really 0.

To be fair: There are lots of options for a headless cms - I'm extremely confident that I could easily move my data to another free or super cheap service - even back to a free wordpress site if need be (to use as headless cms).

Also lots of extremely cheap options for free or very cheap hosting of static sites. That would also be an incredibly easy move to make.

Not trying to fanboy Not saying gatsby is right for every use case and every developer Am saying that it makes a lot of sense for my specific use case and probably for a lot of this agency and old-school-business-website style work. I suspect it is a very large market that provides decent livelihoods for many millenials that skated, snowboarded, surfed, or played in bands as young adults (kind of serious)

I just wanted to share that as an personal experience that people can weigh against other viewpoints.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: