Hacker News Reader

Comment by walshemj | Hacker News Reader

They where saying that in the 1970's how do you automate even basic front end development of a web site today ?

Just like the old gypsy woman said, today a huge percentage of white-collar office work could literally be replaced with a shell script or a couple days' work by someone who knows their way around excel. But it still hasn't happened and attempts to add more software to these jobs seem usually to do very little to improve efficiency, often just adding more jobs to support the software. Every time more software's added, actual automation gets harder, too, as more systems become involved. Pretty soon a couple macros and a script won't cut it, because there are so many pieces involved they'd certainly break all the time.

I'm not even sure most office technology's helping that much. I suspect most of the efficiency gains from tech have been in a handful of "knowledge" work areas, and manufacturing. Plus new markets. The rest... might actually be losing everyone money.

Have you heard of robotic process automation ? It looks promising to replace office people’s jobs .

This is actually kinda what the person you’re replying to is talking about. You have a piece of software that replaces some knowledge workers’ task. At first it seems like you are saving money, but the specifics of the task require an adaptable workforce over the course of years, and the RPA requires an adaptable collection of workers to learn from. Now, instead of just having a job that processes those inputs and outputs, you have the same thing but made more baroque by being connected to RPA, plus software developers to maintain the RPA software itself.

Short term, it seems to eliminate workers. Shortly after those workers are eliminated, truth on the ground has changed, and now there’s an ADDITIONAL cost to updating the software, which is harder and more expensive to do without the workers that have been eliminated.

RPA does eliminate workers. Modern RPA isn’t just “write a screen scraper and never think about it again” — you use BAs and line workers to (cheaply) build out a bunch of RPA integrations, see which ones are the most popular / valuable through usage data from your RPA platform, then implement those via less brittle methods as part of your existing roadmaps. The important part is that the front-end that users access stays the same — you’re basically just swapping an API call to an RPA platform with a direct API call to a backing webservice.

This allows for rapid development of functional integrations delivering real business capabilities — without the overhead of an SDLC. It helps wrangle shadow IT tools into a manageable system and generates lots of structural data on how the actual users of the system interact with it to guide investment in feature development. The RPA scripts with little / no value will naturally fall into disrepair, while the valuable ones will be maintained / improved.

I promise you, it’s way cheaper this way. Software teams stop getting inundated with one-off feature requests that take a ton of work to coordinate, scope, evaluate and test — even if they never get built. Business users get to eliminate time-consuming manual processes. Also keep in mind the first jobs to get replaced by RPA are the ones that have already been shipped offshore... An RPA team for a large enterprise is usually only 2 or 3 engineers and an army of BAs and PMs — but they’re replacing a bunch of individual projects that no longer need to be staffed.

Make a tool that automatically converts a designer's spec/mockup into a performant webpage with high fidelity? Sounds very hard but certainly possible

To be fair there are already a lot of services that do this. They are just so commonplace we don't think of them as automation. And they don't allow super complicated setups so they're definitely not quite there yet in the general case. I'm thinking of Wix, Squarespace, basic wordpress.

Yeah, exactly things like that.

The SaaS marketplace became so crowded that companies sought out a wide range of differentiated niches and automated a lot of specific use cases.

Have you seen the average spec for a website?

A tonne of mostly front end work is now done in tools like wix and even on the higher end wordpress has taken over a lot of it. The only custom front-end web design work tends to be clients with deeper pockets but local web design shops for small business are largely a thing of the past along with the hosting etc.

Aside from the job losses we've lost a gateway into our industry for graduates.

Reply on news.ycombinator.com