7 ms·

I actually did some freelance website development for a local newspaper website around 15 years ago. I can't speak for larger newspapers, this was my experience

I actually did some freelance website development for a local newspaper website around 15 years ago. I can't speak for larger newspapers, this was my experience with a much smaller one. The biggest challenges I found were: 1. Lack of a real technical / engineering department. Having said that, their core business is news and they really just need a functioning website, not a full (and expensive) engineering department. 2. They wanted the website to "feel" more like a real newspaper, including things like incorporating custom layout, design and other elements on the website. Explaining the limitations of HTML, CSS and Javascript was a bit ... frustrating for both sides. 3. Too many opinions on what the website should look like. Ditto for even small features such as what the photo gallery and subscription pages should look like. 4. Too many feature requests, often contradicting each other. It wasn't clear who was in charge of making decisions on website changes.

expertentipp 8y ago
> They wanted the website to "feel" more like a real newspaper, including things like incorporating custom layout, design and other elements on the website. Explaining the limitations of HTML, CSS and Javascript was a bit ... frustrating for both sides. Did they demand "page turning animation" each time user clicks "next page" in the pagination?
- gameswithgo 8y ago
  I think 15 years ago the main misunderstanding would be that web pages have to able to flow around the dimensions of the window that the user has chosen. You can't fix things in particular places as newspaper people would be used to. I imagine today people are more familiar with that concept.
  - annabellish 8y ago
    >I imagine today people are more familiar with that concept. You... would be surprised. I've been asked to implement "pixel perfect" designs on mid-six figure projects, and explaining that that isn't how the web works can be fruitless when their background is in traditional media.
    - expertentipp 8y ago
      Oh boy, the "pixel perfect" storm period couple of years ago. Luckily I waiter it over on cozy position without a need of looking for a job.
- coldtea 8y ago
  In the early days of the web, lots of print-originating online outlets, did.
zebraflask 8y ago
Good points. I'd expand that by saying that these apply to any organization that wants a content-based site but has no need for a real web development team. Non-technical users often don't know what they want until they see something on the screen, and once they see it on the screen, they inevitably want to make changes that undo the first set of things they asked for. Everybody seems to have an opinion about layouts and design features, whether they ought to or not, and the least-capable voices tend to be the loudest. And you're absolutely right, explaining the limits of web dev to non-technical people is a perpetual exercise in frustration. They think web dev is like print design. As we know, it's not even close. Web design is constrained by a multitude of browser limitations that print simply does not have. They don't like to hear that.
- hobs 8y ago
  Often considered "bike shedding" https://phinze.github.io/2014/05/24/useful-tech-terms-part-1.html https://phinze.github.io/2014/05/24/useful-tech-terms-part-1... When you are working on minor details, everyone has an opinion. When you are working on major difficult things, nobody wants to weigh in.
- qznc 8y ago
  > explaining the limits of web dev to non-technical people is a perpetual exercise in frustration That is doing the wrong thing. These non-technical people don't need to understand "limits". They need someone with a clear process who guides them through it. Unfortunately many web devs have no process. Print design has its own limitations but the process which people have to follow is more established.
  - zebraflask 8y ago
    Sometimes that works, but as a counterpoint: if your stakeholders outrank you in the org and don't care for your process or choose to ignore it, good luck!
- paultopia 8y ago
  > Non-technical users often don't know what they want until they see something on the screen, and once they see it on the screen, they inevitably want to make changes that undo the first set of things they asked for. Everybody seems to have an opinion about layouts and design features, whether they ought to or not, and the least-capable voices tend to be the loudest. What does that have to do with technical or non-technical? Isn't that just visual vs non-visual? Print designers complain about the same thing. Personally, as someone who is technical but not visual, I know I definitely don't know what I want when putting together anything visual until I build it, because I can't form a clear enough picture in my mind of what the end product will look like. So I inevitably end up building something, being really really miserable about it, and then tweaking it until I'm only moderately sad about it.
ryandrake 8y ago
> 1. Lack of a real technical / engineering department. Having said that, their core business is news and they really just need a functioning website, not a full (and expensive) engineering department. I agree with the rest of your points but this one doesn't really make sense. My guess is it would take fewer engineers to make these websites better. The horrible parts of newspaper websites are the pop-ups and broken scrolling and ads that follow you and fixed-position "dickbars" at the top and bottom--things that don't come with simple HTML and need to be deliberately added by developers. Newspapers really do just need a functioning website. Instead they pay developers to add in all these horrible things.
- jasoncartwright 8y ago
  Slight aside: Django came out of a local newspaper https://en.wikipedia.org/wiki/Django_(web_framework)#History https://en.wikipedia.org/wiki/Django_(web_framework)#History
  - knowtheory 8y ago
    Right, and it's still sad to say that they capitalized on that fact so incredibly poorly that the following came to pass: https://twitter.com/jacobian/status/1012781017940316161 https://twitter.com/jacobian/status/1012781017940316161 > The Lawrence Journal-World, where Django was created, is now a Wordpress site. http://www2.ljworld.com/news/2018/jun/26/redesign-ljworld/ http://www2.ljworld.com/news/2018/jun/26/redesign-ljworld/
    - always_good 8y ago
      Or they realized that there isn't much value in building their own CMS.
      - ubernostrum 8y ago
        I worked at the Journal-World for around five years, starting in 2006. Upper management at the time was interesting. A lot of people who didn't necessarily understand the internet or technology in general, but knew they didn't, and were willing to hire and trust people who did. That was the magic sauce that led to Django, and a lot of the other innovative stuff. The owner of the paper, for example, had his secretary print out his emails and bring them into his office; then he'd write his replies on a manual typewriter, and hand them back to be typed into a computer. But he'd also managed to ride the wave of first the cable TV/internet boom (by setting up a cable division) and then the web boom (by hiring a team of people to build a first-class news site and giving them more or less free reign to do it right). And that was how you got the heyday of the Journal-World. All sorts of interesting experiments in using the web to enhance journalism, close collaboration between the newsroom and the tech team, and a ton of cool things accomplished and a bunch of industry awards, etc. I actually had a byline at one point, on a feature that's now gone because Django got retired (a data-journalism project tracking the impact of flu during the H1N1 scare). And other news organizations were happy to pay for the software to do their own version of that. We had both hosted and on-prem versions of it, and recommendations for hiring developers and training them to work on it, and as far as I could tell they seemed pretty happy to have something that had been designed at and by a newspaper (as opposed to other news CMS products, which often have their first encounter with a journalist at the time of production deployment). But all good things come to an end. There were some management shakeups, and a lot of the tech team (myself included) left for greener pastures. A little while after that, I heard the CMS division was being shut down and everyone in it laid off; then I heard another company had made an offer to acquire it. As far as I know, Ellington (the Django-based news CMS) is still available today as a supported commercial product. But the Journal-World no longer uses it or, to my knowledge, maintains an in-house technology team like it used to.
ravenstine 8y ago
> Too many feature requests, often contradicting each other. My last job was at a public radio station that published news online. That was one of my biggest frustrations, in part because it takes time away from properly addressing bugs or investigating what features users don't want.
forapurpose 8y ago
Many of these issues point to the same root cause: A vacuum of leadership at the executive level. Those people often don't understand the Internet, UX, etc., and thus can't and don't make good decisions, creating chaos beneath them. For example, until recently at the New York Times, the head of social media did not use social media himself.[0] Think about that - think about how important social media is to the NYT. Imagine any competent company - much less a world leader - hiring for that position; they probably have their choice of applicants and that's who they choose? The bad news is that he no longer runs social media at the NYT, he runs the company - A.G. Sulzberger. How will the NYT and its technical side fare under that leadership? (That ignores the deep blow to morale of such nepotism as making him head of social media or, in times like this, the company - he's a scion of the family that has run the NYT for generations.) [0] https://www.wired.com/2017/02/new-york-times-digital-journalism/ https://www.wired.com/2017/02/new-york-times-digital-journal...
- ACow_Adonis 8y ago
  To be fair, it's generally accepted knowledge that drug dealers and producers probably shouldn't be consuming their own product.
microcolonel 8y ago
I worked with a major news network on a UI prototype/hack which effectively destroyed and remounted the entire DOM after load.... and they put it into production as-is.
qwerty456127 8y ago
> Lack of a real technical / engineering department A decent news website is quite easy to set up with Jekyll or WordPress and there are so many people that can do this around all over the world, there is no real need for an engineering department.
- firebones 8y ago
  What's an example of a good news website that uses Jekyll or WordPress?
  - CharlesW 8y ago
    A "good" (which is subjective) news site isn't guaranteed or prevented by the use of any particular CMS. A quick Google search shows that many popular news sites are WordPress-powered, including BBC America, The New York Times, The New Yorker, Quartz, TechCrunch, Time, and Variety. Many of these are considered good.
  - qwerty456127 8y ago
    I didn't say good. I said decent. As for me I would avoid WordPress at all costs but the fact is it can be used to get the thing done tolerably well and there are just so many people who can operate it.
BurningFrog 8y ago
This sounds utterly ripe for someone to develop a standard newspaper web site that all the thousands of newspapers around the world can adopt with some minimal branding and customization. I assume/hope someone did this in the 15 years since?
- jumelles 8y ago
  I don't think they ever will, but it would be incredible if the New York Times open-sourced their tools. https://open.nytimes.com/building-a-text-editor-for-a-digital-first-newsroom-f1cb8367fc21 https://open.nytimes.com/building-a-text-editor-for-a-digita... https://open.blogs.nytimes.com/2014/06/17/scoop-a-glimpse-into-the-nytimes-cms/ https://open.blogs.nytimes.com/2014/06/17/scoop-a-glimpse-in...
  - twsted 8y ago
    Oak is based on the open source and very good ProseMirror.
- jinglebells 8y ago
  Like The Guardian? https://github.com/guardian/frontend https://github.com/guardian/frontend
BurningFrog 8y ago
> Too many feature requests, often contradicting each other. It wasn't clear who was in charge of making decisions on website changes. This is the default state of every software project that doesn't have an unusually mature customer. It is your job to try to be an unusually mature supplier and make these tradeoffs explicit to them.
KuhlMensch 8y ago
15 years ago is a long time ago.
kminehart 8y ago
I work for a news company right now. Or rather, the parent company of a local news, similar to Gannett (the company that was mentioned in the article). I've found that we have a lot of freedom, and we've recently deployed a redesigned website that we're happy with. There's definitely been a lot of frustration from arguing with all the different departments: advertising, circulation, editorial / the newsroom, but the most frustrating aspect by far has been the technical debt we still have yet to address properly. Software that was purchased by our company several years ago for managing subscribers and their subscriptions is abysmal. It's used far beyond its scope, where generic "notes" fields are used to store critical information. For example, these systems are for print subscriptions only, so they amortize billing, and there's no email address field. We've shoehorned online subscription functionality by including email addresses in the notes field. There's no API or any sort of way to truly interact with these "systems" either, so customers are just flat-out unable to manage their own account without some middle-man manually entering form-submitted data. Truthfully I feel that even with our redesigned website (which has gotten overwhelmingly positive reviews), we still suffer from an awful online presence. What sort of subscription service exists these days that doesn't even let you change your address without calling? It's pretty sad.

expertentipp 8y ago

> They wanted the website to "feel" more like a real newspaper, including things like incorporating custom layout, design and other elements on the website. Explaining the limitations of HTML, CSS and Javascript was a bit ... frustrating for both sides. Did they demand "page turning animation" each time user clicks "next page" in the pagination?

gameswithgo 8y ago
I think 15 years ago the main misunderstanding would be that web pages have to able to flow around the dimensions of the window that the user has chosen. You can't fix things in particular places as newspaper people would be used to. I imagine today people are more familiar with that concept.
- annabellish 8y ago
  >I imagine today people are more familiar with that concept. You... would be surprised. I've been asked to implement "pixel perfect" designs on mid-six figure projects, and explaining that that isn't how the web works can be fruitless when their background is in traditional media.
  - expertentipp 8y ago
    Oh boy, the "pixel perfect" storm period couple of years ago. Luckily I waiter it over on cozy position without a need of looking for a job.
coldtea 8y ago
In the early days of the web, lots of print-originating online outlets, did.

zebraflask 8y ago

Good points. I'd expand that by saying that these apply to any organization that wants a content-based site but has no need for a real web development team. Non-technical users often don't know what they want until they see something on the screen, and once they see it on the screen, they inevitably want to make changes that undo the first set of things they asked for. Everybody seems to have an opinion about layouts and design features, whether they ought to or not, and the least-capable voices tend to be the loudest. And you're absolutely right, explaining the limits of web dev to non-technical people is a perpetual exercise in frustration. They think web dev is like print design. As we know, it's not even close. Web design is constrained by a multitude of browser limitations that print simply does not have. They don't like to hear that.

hobs 8y ago
Often considered "bike shedding" https://phinze.github.io/2014/05/24/useful-tech-terms-part-1.html https://phinze.github.io/2014/05/24/useful-tech-terms-part-1... When you are working on minor details, everyone has an opinion. When you are working on major difficult things, nobody wants to weigh in.
qznc 8y ago
> explaining the limits of web dev to non-technical people is a perpetual exercise in frustration That is doing the wrong thing. These non-technical people don't need to understand "limits". They need someone with a clear process who guides them through it. Unfortunately many web devs have no process. Print design has its own limitations but the process which people have to follow is more established.
- zebraflask 8y ago
  Sometimes that works, but as a counterpoint: if your stakeholders outrank you in the org and don't care for your process or choose to ignore it, good luck!
paultopia 8y ago
> Non-technical users often don't know what they want until they see something on the screen, and once they see it on the screen, they inevitably want to make changes that undo the first set of things they asked for. Everybody seems to have an opinion about layouts and design features, whether they ought to or not, and the least-capable voices tend to be the loudest. What does that have to do with technical or non-technical? Isn't that just visual vs non-visual? Print designers complain about the same thing. Personally, as someone who is technical but not visual, I know I definitely don't know what I want when putting together anything visual until I build it, because I can't form a clear enough picture in my mind of what the end product will look like. So I inevitably end up building something, being really really miserable about it, and then tweaking it until I'm only moderately sad about it.

ryandrake 8y ago

> 1. Lack of a real technical / engineering department. Having said that, their core business is news and they really just need a functioning website, not a full (and expensive) engineering department. I agree with the rest of your points but this one doesn't really make sense. My guess is it would take fewer engineers to make these websites better. The horrible parts of newspaper websites are the pop-ups and broken scrolling and ads that follow you and fixed-position "dickbars" at the top and bottom--things that don't come with simple HTML and need to be deliberately added by developers. Newspapers really do just need a functioning website. Instead they pay developers to add in all these horrible things.

jasoncartwright 8y ago
Slight aside: Django came out of a local newspaper https://en.wikipedia.org/wiki/Django_(web_framework)#History https://en.wikipedia.org/wiki/Django_(web_framework)#History
- knowtheory 8y ago
  Right, and it's still sad to say that they capitalized on that fact so incredibly poorly that the following came to pass: https://twitter.com/jacobian/status/1012781017940316161 https://twitter.com/jacobian/status/1012781017940316161 > The Lawrence Journal-World, where Django was created, is now a Wordpress site. http://www2.ljworld.com/news/2018/jun/26/redesign-ljworld/ http://www2.ljworld.com/news/2018/jun/26/redesign-ljworld/
  - always_good 8y ago
    Or they realized that there isn't much value in building their own CMS.
    - ubernostrum 8y ago
      I worked at the Journal-World for around five years, starting in 2006. Upper management at the time was interesting. A lot of people who didn't necessarily understand the internet or technology in general, but knew they didn't, and were willing to hire and trust people who did. That was the magic sauce that led to Django, and a lot of the other innovative stuff. The owner of the paper, for example, had his secretary print out his emails and bring them into his office; then he'd write his replies on a manual typewriter, and hand them back to be typed into a computer. But he'd also managed to ride the wave of first the cable TV/internet boom (by setting up a cable division) and then the web boom (by hiring a team of people to build a first-class news site and giving them more or less free reign to do it right). And that was how you got the heyday of the Journal-World. All sorts of interesting experiments in using the web to enhance journalism, close collaboration between the newsroom and the tech team, and a ton of cool things accomplished and a bunch of industry awards, etc. I actually had a byline at one point, on a feature that's now gone because Django got retired (a data-journalism project tracking the impact of flu during the H1N1 scare). And other news organizations were happy to pay for the software to do their own version of that. We had both hosted and on-prem versions of it, and recommendations for hiring developers and training them to work on it, and as far as I could tell they seemed pretty happy to have something that had been designed at and by a newspaper (as opposed to other news CMS products, which often have their first encounter with a journalist at the time of production deployment). But all good things come to an end. There were some management shakeups, and a lot of the tech team (myself included) left for greener pastures. A little while after that, I heard the CMS division was being shut down and everyone in it laid off; then I heard another company had made an offer to acquire it. As far as I know, Ellington (the Django-based news CMS) is still available today as a supported commercial product. But the Journal-World no longer uses it or, to my knowledge, maintains an in-house technology team like it used to.

ravenstine 8y ago

> Too many feature requests, often contradicting each other. My last job was at a public radio station that published news online. That was one of my biggest frustrations, in part because it takes time away from properly addressing bugs or investigating what features users don't want.

forapurpose 8y ago

Many of these issues point to the same root cause: A vacuum of leadership at the executive level. Those people often don't understand the Internet, UX, etc., and thus can't and don't make good decisions, creating chaos beneath them. For example, until recently at the New York Times, the head of social media did not use social media himself.[0] Think about that - think about how important social media is to the NYT. Imagine any competent company - much less a world leader - hiring for that position; they probably have their choice of applicants and that's who they choose? The bad news is that he no longer runs social media at the NYT, he runs the company - A.G. Sulzberger. How will the NYT and its technical side fare under that leadership? (That ignores the deep blow to morale of such nepotism as making him head of social media or, in times like this, the company - he's a scion of the family that has run the NYT for generations.) [0] https://www.wired.com/2017/02/new-york-times-digital-journalism/ https://www.wired.com/2017/02/new-york-times-digital-journal...

ACow_Adonis 8y ago
To be fair, it's generally accepted knowledge that drug dealers and producers probably shouldn't be consuming their own product.

microcolonel 8y ago

I worked with a major news network on a UI prototype/hack which effectively destroyed and remounted the entire DOM after load.... and they put it into production as-is.

qwerty456127 8y ago

> Lack of a real technical / engineering department A decent news website is quite easy to set up with Jekyll or WordPress and there are so many people that can do this around all over the world, there is no real need for an engineering department.

firebones 8y ago
What's an example of a good news website that uses Jekyll or WordPress?
- CharlesW 8y ago
  A "good" (which is subjective) news site isn't guaranteed or prevented by the use of any particular CMS. A quick Google search shows that many popular news sites are WordPress-powered, including BBC America, The New York Times, The New Yorker, Quartz, TechCrunch, Time, and Variety. Many of these are considered good.
- qwerty456127 8y ago
  I didn't say good. I said decent. As for me I would avoid WordPress at all costs but the fact is it can be used to get the thing done tolerably well and there are just so many people who can operate it.

BurningFrog 8y ago

This sounds utterly ripe for someone to develop a standard newspaper web site that all the thousands of newspapers around the world can adopt with some minimal branding and customization. I assume/hope someone did this in the 15 years since?

jumelles 8y ago
I don't think they ever will, but it would be incredible if the New York Times open-sourced their tools. https://open.nytimes.com/building-a-text-editor-for-a-digital-first-newsroom-f1cb8367fc21 https://open.nytimes.com/building-a-text-editor-for-a-digita... https://open.blogs.nytimes.com/2014/06/17/scoop-a-glimpse-into-the-nytimes-cms/ https://open.blogs.nytimes.com/2014/06/17/scoop-a-glimpse-in...
- twsted 8y ago
  Oak is based on the open source and very good ProseMirror.
jinglebells 8y ago
Like The Guardian? https://github.com/guardian/frontend https://github.com/guardian/frontend

BurningFrog 8y ago

> Too many feature requests, often contradicting each other. It wasn't clear who was in charge of making decisions on website changes. This is the default state of every software project that doesn't have an unusually mature customer. It is your job to try to be an unusually mature supplier and make these tradeoffs explicit to them.

KuhlMensch 8y ago

15 years ago is a long time ago.

kminehart 8y ago

I work for a news company right now. Or rather, the parent company of a local news, similar to Gannett (the company that was mentioned in the article). I've found that we have a lot of freedom, and we've recently deployed a redesigned website that we're happy with. There's definitely been a lot of frustration from arguing with all the different departments: advertising, circulation, editorial / the newsroom, but the most frustrating aspect by far has been the technical debt we still have yet to address properly. Software that was purchased by our company several years ago for managing subscribers and their subscriptions is abysmal. It's used far beyond its scope, where generic "notes" fields are used to store critical information. For example, these systems are for print subscriptions only, so they amortize billing, and there's no email address field. We've shoehorned online subscription functionality by including email addresses in the notes field. There's no API or any sort of way to truly interact with these "systems" either, so customers are just flat-out unable to manage their own account without some middle-man manually entering form-submitted data. Truthfully I feel that even with our redesigned website (which has gotten overwhelmingly positive reviews), we still suffer from an awful online presence. What sort of subscription service exists these days that doesn't even let you change your address without calling? It's pretty sad.