I get your point, and I think it gets to the heart of the matter.
Now that everyone uses Facebook, Gmail, Amazon, Twitter on a daily basis, the average person thinks of those as simple 'websites', that are easy to build.
Almost daily I have friends/family come to me with a business idea that starts off as "it will be easy.. just like <huge site> but with blah".
Non-tech people honestly think 'websites' are simple things, and we should be able to knock-out a Facebook or Twitter clone in a few days.
It wasn't that long ago that everyone had a 13 year old nephew that made websites. Also it is easier to create a very simple website than to create a very simple desktop app, since a simple website doesn't really do anything. There's really no equivalent desktop app to the Geocities type websites.
> Now that everyone uses Facebook, Gmail, Amazon, Twitter on a daily basis, the average person thinks of those as simple 'websites', that are easy to build.
I'm really surprised that people actually look at the websites of multi-million dollar internet-only companies and think it must be easy to build something worth that much.
Why does it matter to the average Joe how much a random Tech company is worth? at the end of the day, it's all made up anyway, and the vast majority are waaaaay to concerned with their next payment to think about that stuff anyway.
I'm a developer so I know how to do these things but for the average person...
Where do I type that?
By the way.. What is apt-get? What is nginx?
Why put the index.html in /var/www?
Why use nginx I've heard of apache why wouldn't I use that?
You haven't even detailed how to setup a linux that would use apt. And why not centos?
Where is the site going to be hosted, AWS/Digital Ocean/ Linode/ Heroku? How do they properly configure the machine to serve to internet?
How are my non-technical clients/employees/founders/etc. going to update it without bugging me all the time?
In other words my friend you've grossly over simplified the process of putting a website on the internet where other people can get to it and making it easily updatable by laymen
Exactly, no average person will be questioning the choice of apache over nginx or vice-versa. Nor will they be evaluating choice of hosting provider. The average person probably has no clue what AWS is.
Instead the average person can be expected to google for something along the lines of "free website" or something similar, arrive at wordpress/blogger/tumblr/squarespace/flavour-of-month .com and just create an account.
And that's where the divide resides. Some people take great pleasure in doing everything from scratch and having endless configurability. Everyone else wants to throw something together and push a button to let the world use it.
I'd like to build just a website and have the rest set to an automatic that I really don't care about configuring.
The premise, for non-technical is perfect. I've heard stakeholders and clients saying that making a website is like pushing a button.
The project itself seems interesting nonetheless. I just need to see the limitations of it.
As nginx and others can be quite powerful.
"These days, the person doing the development is often different from the person doing the visual design, is different from the person structuring the content, is different from the person actually writing the content, is different from the person reviewing the launch, is different from the person writing the check to pay for the whole thing"
...uhhh, not unless you work in a bloated enterprise shop. Most of the time, it's a one to three person show and one person CAN do it all (albeit maybe not everything equally perfect). I launch websites (whole platforms actually) - from server to web services to content - all the time, every few days. It is entirely possible and you get accustomed to it.
Err, I'd say that it's getting easier then ever for one person to build a website by themselves.
- Rails/Angular go a long way to speeding up development due to the number of 3rd party libraries available for them. Need User logins? Devise. Need queue processing? Sidekiq. AngularJS is going the same way as Rails for awesome 3rd party libraries.
- Bootstrap almost solves visual design for prototypes, and it's not hard to start customising the templates or even buying a 3rd party one off the shelf. It's not going to win any awards for graphic design but it's fine for 99% of people that will be using your site.
Case in point, I built http://www.knowyourgenre.com, as site for exploring music by genres, in around 3 months (with a lot of backtracking and taking bits out that I didn't wind up using).
1 person, I did the front end, back end and graphic design for it.
Not to say that grow is a bad idea, but I don't agree with the argument that 1 person can't build a site anymore, in fact I think it's getting easier.
Thanks for putting that into words more nicely than I would have been able to. Grow looks interesting and I applaud them for the initiative, but it's hard to agree with all that they are saying when their definition of a "web page" is not at all congruent with my personal definition.
For anyone at Grow that is reading this - I think you might find more people in agreement if you show them an apples to apples comparison. Show me all that it would take to launch comparable apps on GAE vs. Grow and let that speak for itself.
OP here. I totally agree with your statement in that it's becoming easier and easier for single developers to build (and launch) something cool.
I suppose I must have glanced over this when I wrote the post, but I'm definitely targeting Grow towards producing content-heavy informational marketing web sites, blogs, portfolios, landing pages, personal sites, corporate information pages, marketing campaigns, etc.
Perhaps I'm not your target audience but I would precisely avoid your website because of the poor, clearly bootstrapped, design.
Other than that I completely agree with you that it's easier now more than ever to get a site up and running. Design however takes more thought than throwing some existing bootstrap styles on there.
Ok, stepping back and critiquing here, the call to action falls hard on the fact that 'website' doesn't have a common enough definition.
It is in fact trivial to build a web site, one person. My sister did it (not technical) went to WPEngine.com got the basic package, went to name.com and got a domain that wasn't taken yet, and added Google apps for the email component. Boom, "web site" and she is very non-technical, but she does know how to use a search engine.
Then there are the Geo/NeoCitites type things - also pretty trivial. And new entrant Ghost, also easy to do for one person.
Then there are the Github Pages, Dropbox pages, and Google Drive pages (and Google Sites for that matter, an unwanted love child with Jot).
If you are actually modestly technical then it gets even easier.
So that whole "its rocket science" meme doesn't really fly and that takes away from the whole 'this is hard we're here to help' theme.
There are tons of website builders that allow you to create simple (to rather complex) websites nowadays. Squarespace, Jimdo, Wix to name a few can be great tools depending on your needs. No technical skills required, content creation and publication tend to be pretty accessible and all the hosting and domain name handling is usually fully taken care of.
I'm myself working for a French startup called Pikock[1] that is working hard on the next iteration of its drag-and-drop website builder. We're trying to make it as simple, efficient and streamlined as possible so that everyone gets a shot beyond the tech-savvy. Our core vision is that everyone, no matter the skill level is entitled to a presence online if they so choose.
Ultimately I want my 65-year old mom to be able to create content and share it with the world, even though there might not be an audience for her collection of hand-painted pebbles :)
Yeah I'm always surprised about how hard it is to put a website online.
I think there's a market for a service helping a team getting started on the right foot.
It could start with a popular stack of services and frameworks, let's say:
- Github, Heroku, S3, node.js, angular, grunt, bower, yo, bootstrap, etc etc.
And just make it dead simple to have it configured and deployed.
Now, all these services are awesome by themselves, but they all require their own quirks and quacks to work together which makes it a real pain to get started.
Oh, and it should have a nice menu system for doing things related to the site when on a page.
Oh, and it should respond to the device and browser of the person using it, with the layout and functionality adjusting accordingly.
Oh, and the images should be the appropriate size for that device, because we don't want users on cellular networks to have to download large images.
Oh, and it should display for them in their preferred language. And without having to choose the language from a menu. Don't tell me about this internationali-whatsit and local-whoozits, I said it should display in their language! (God, you developers are annoying.)
Oh, and it should mirror the smoothness of the native applications of the user's device.
Oh, and it should be built on roughly the same technologies which websites have been using for the last 15 years. And I don't want to hear about "HTML is a system for describing documents, not applications" or "CSS is for styling documents, it's not a layout engine".
Yeah, you know, jeez, why is this so hard????? Everyone is stupid and in my way.
Coming up with an algorithm is not hard. It is just time consuming and annoying because brute force is stupid to people. So, gluing code in some brain dead fashion (wtf do I do here? Oh no better solution so I will patch this patch that with stitches) is hard because it is like brutcing. There are too many "libraries" and none can do everything easily. Complex system is the end result. Then engineers spend most of their time reducing the complexity. If there was a framework that could reduce in the first stage then it is easier, not harder.
I thought this was going to be an argument for simplicity but then it started talking about complex things. Git and SASS and Python and whatnot are great, but if you just want to build a website, like in the Good Old Days, surely https://neocities.org/ does the trick. Put HTML files in your 10MB of space.
This sounds like a friendlier Heroku and may well turn out to be brilliant, but it is still a lot more than just building a website.
When I made our wedding website, I wrote a short python cgi script to push markdown files through a parser and paste them into a template. Then I wrote a CSS file to make it pretty, and popped that up with a static HTML home page on NearlyFreeSpeech behind CloudFlare. We're talking about 27 lines of python, 62 lines of HTML, and 135 lines of CSS.
My point here isn't that people should do exactly what I did. Rather, my point is that most of a web developer's job is finding the right components and then writing a little bit of code to glue them together. Stop looking for some special system that holds your hand like a toddler through every step of every process, and just write some damned code to do the thing you need done.
I personally don't believe in learning and deploying an entire framework to save less than 250 lines of code.
This is precisely why web-hosting providers exist. You can find someone to host most of the common CMSs, you can certainly find a company that will run a web server on your behalf (so you can deploy a static site). If you really want a tuned WordPress installation there's WPEngine.
If you don't want to deal with the installation and configuration of the underlying software, don't lease a VPS or any other type of server ... lease a "service".
EDIT: I completely forgot about sites like Weebly. You don't even have any research to do with this type of solution as you just use what you're given.
I'm working on something that's sort of the opposite called Strapper[1]. The thing is, when I'm trying to launch a new product, I don't even want to build a website. It's a huge distraction. And I don't want to build on a CMS like Wordpress, because I'll just have to replace it with Rails when the project scope grows.
Different strokes for different folks though. What someone needs to build a blog or portfolio is going to be different from what you need to build a startup.
I think your messaging could be a bit more clear. What you do is pretty abstract. After reading the article and your landing page, I'm not actually sure what your product does or why I would want it.
If someone doesn't like the "modern" way of building websites, or thinks that it requires so much more work than the old way - then I recommend they try the old way.
Announces a website launcher thing that deemphasizes configuration without preference for dynamic/static or local/cloud. (Not sure how precise that is. Says site will be file-based and has browser site editor.) Mentions capable with AWS and Google cloud. Allows collaboration. Launch can be specified for a certain date. User controls. And git is the vcs.
I find the writeup similarly misleading.
"all I want to do is build a web site"...
"... hosted on google apps..".
... manage content
... collaborate among team members
... CMSy stuff.
So in other words, all you want to do is build a pipelined product. Which is totally cool - but it's definitely not just building a web site.