The cost of adding features (and why we say no to most suggestions)

Matt M

Director Development at Invision Community
Joined
Apr 28, 2005
Messages
257
I wrote a few words about why we say "no" to most feature ideas, and how we process feature suggestions internally.


Do you think we're right? Or should we just add whatever customers ask us to?
 

haqzore

Devotee
Joined
Dec 6, 2012
Messages
2,327
I think the approach is, well, smart. It's that simple.

But as I've frequently stated, I hate the fear of 3rd party add-ons being abandoned. There should be therapy for it. And when something I want isn't in the core, third party solutions are the only option.
 

zappaDPJ

Administrator
Joined
Aug 26, 2010
Messages
7,232
Do you think we're right? Or should we just add whatever customers ask us to?
I think you must be right because from where I stand Invision Community is ticking far more boxes than the competition. I would however like to hear the other half of the story; the decision process that leads to added features such as the native app but perhaps that's a blog for another day :)
 

sbjsbj

Fan
Joined
Feb 9, 2015
Messages
663
It is a good read, first of all.

Can you explain this further?
Our biggest misstep was spending 18 months creating a new application that would take us into a brand new market. We eventually pulled it completely, but that's another story.

About the whole article:
Obviously I, as a customer, will demand features. Obviously you, as a developer, will say no to the requests.

I understand that it is impossible to implement everything, but I hate particularly this approach:
We never half-ass something, but we frequently half-implement it.
Yep, that is why all of us have to rely on addons this much. You guys either knowingly do this (as you confess) or you guys have no feeling for your userbase and don't know what we want.

In both cases it sucks.

If we pay a monthly, semi-yearly, yearly pay, then we expect enhancements. Easy as that. If I pay Netflix 10 bucks a month, I expect something in return. I expect the newest movies or series in the library. I don't care about what was 10 years ago.
If I pay a forum software company regularly money, I expect it to be on par with the modern times. If you guys still lack 10 years, then there is something wrong.
Modern development evolves much faster. The arrogant attitude of saying "no" does get you what? You are not even 0.001% relevant to modern userbase. Nobody cares. Is this what you want? Is this your strategy?

Now of course I sound a bit harsh but it gets the point across.

I am saying this but I really really really want you to be successful. I gain nothing from hating or that you guys lack stuff. I want you to be popular and awesome and the best. But you on the steering wheel don't have the same enthusiasm as we do or as some addon developers do. At least this is what I feel in the whole forum world. You are coasting and melking us. After 15 years if vB3 still does 80% the same job as the other ones, would you take a minute and ask yourself what you did in the last 15 years? What have you invented that vB3 didn't have? What innovation? There is a good reason why still so many communities are on vB3 or vB4. The tempo you bring to this modern world doesn't cut it for the customer, so we will demand more and more.

PS: I didn't want to sound this harsh but as you took the liberty to say "no". I thought I might aswell take the liberty to say "no" to your "no". :) But all in all it was a good read, wish our devs put out articles like that. No offense intented and with all respect. Keep posting such articles and don't let replies such as mine stop you from doing that. It shows enthusiasm to share such articles with the community and it is a good sign, I like that.
 
Last edited:

Matt M

Director Development at Invision Community
Joined
Apr 28, 2005
Messages
257
Well, there's a lot to unpick here. :D

"Obviously I, as a customer, will demand features. Obviously you, as a developer, will say no to the requests."

If you read the whole article, that's not strictly true. We just don't act on every request that comes in. We sometimes watch and wait to see if a single suggestion forms part of a trend of suggestions.

So, we don't "ignore" feedback at all. As mentioned a few times in there, we welcome it and appreciate it.

"Yep, that is why all of us have to rely on addons this much. You guys either knowingly do this (as you confess) or you guys have no feeling for your userbase and don't know what we want."

We absolutely LOVE writing new code and features. However, the truth is that there's only so much time. We're currently looking to wrap up 4.5 and one of the features I'm working on right now has massive potential. To code in ALL the ideas I want to code in would probably take 6 months, which would mean delaying its release until late summer.

Now, that's way too late for the timeframe that we wanted to wrap up this release. The sensible thing is to code in the essentials, and release. Then add more, then release again. And so on.

We do this quite often.

As noted in the article, sometimes our ideas for a feature are dropped if it's not popular (thus putting that dev time elsewhere) or our customers find a different way to use the feature and want to take it in another direction.

Either way, the smart thing to do is to take an iterative approach. We code, we ship, goto 10.

"If I pay a forum software company regularly money, I expect it to be on par with the modern times. If you guys still lack 10 years, then there is something wrong."

I'm not sure what your point is here. You buy a product and pay a renewal. Our renewal is currently $50/year. If you wanted to hire a team of developers to write software just for you, I'd look at having around $250,000 or so to pay for it.

We develop software with the 80% in mind. This means that we know we're not going to make 100% of people happy, so we aim for 80%. We're creating software for hundreds of thousands of brands, business and individuals. Everyone has different priorities.

If you look at our product from 2010 and look at it now, you'll see how massively different it is. I actually wrote a blog on this recently.

"Modern development evolves much faster. The arrogant attitude of saying "no" does get you what? You are not even 0.001% relevant to modern userbase. Nobody cares. Is this what you want? Is this your strategy?"

The reason we say no, and develop essential features when we launch a new feature is so we can develop and ship faster.

Are you claiming we're not relevant to 99.999% of contemporary internet users? That no one cares about our product? If so, you're mistaken :D

Brands like LEGO, Squarespace, Warner Bros., Trip Advisor, SEGA, Sony, NBC, and more have launched communities with Invision Community. This isn't by accident. :)

"But you on the steering wheel don't have the same enthusiasm as we do or as some addon developers do. At least this is what I feel in the whole forum world. You are coasting and melking us. After 15 years if vB3 still does 80% the same job as the other ones, would you take a minute and ask yourself what you did in the last 15 years? What have you invented that vB3 didn't have? What innovation? There is a good reason why still so many communities are on vB3 or vB4. The tempo you bring to this modern world doesn't cut it for the customer, so we will demand more and more."

I can't speak for other platforms. I recently worked out that since we launched Invision Community 4, we have made something like 650 releases. Take a look on our blog and see what we've added recently. You might be surprised. :)

We are not coasting or milking anyone. As mentioned above, our self-host license gives you incredible value in terms of feature set, support and future updates.

We're currently finalising mobile apps for Invision Community, we have recently talked about features we've added to increase engagement, improve retention, increase guest to member ratios and so on.

Communities that are on vB 3 and 4 probably will be for as long as possible because vB3 and 4 is "not even 0.001% relevant to modern userbase" so to switch them to a more modern platform like ours ( because Modern development evolves much faster) would likely cause a revolt and they'd lose their core members without a careful management plan.

I appreciate your reply :)
 

Matt M

Director Development at Invision Community
Joined
Apr 28, 2005
Messages
257
I think you must be right because from where I stand Invision Community is ticking far more boxes than the competition. I would however like to hear the other half of the story; the decision process that leads to added features such as the native app but perhaps that's a blog for another day :)
Sure, I can write that up too! We use a locked down Invision Community as an internal site, with some custom apps to track it all.
 

cheat_master30

Moderator
Joined
Jan 16, 2010
Messages
3,855
Interesting read. Makes sense you wouldn't add most suggested features, especially given how often it is that users don't really know what they want. Or have needs that don't align with the majority of your audience.

Regardless, could you give a bit more insight into this story next? I'm really curious to know what Invision product you scrapped after 18 months in development, or how that situation came about in the first place:

Our biggest misstep was spending 18 months creating a new application that would take us into a brand new market. We eventually pulled it completely, but that's another story.
My first thought was that it was that ecommerce addon you were working on a while back, but it seems that one is still in development.
 

Matt M

Director Development at Invision Community
Joined
Apr 28, 2005
Messages
257
"Regardless, could you give a bit more insight into this story next?"

Yep! I'm aiming to get it written up next week :)
 

overcast

Adherent
Joined
Mar 17, 2019
Messages
475
Loved reading it.
On similar topic, I recently realized how WordPress project got bloated because bunch of javascript developers hijacked it and trying to add every random feature as fancy and trendy into it. I am guessing Gutenberg Block bloat would be one of those which will slowly kill WordPress.
 

MagicalAzareal

Magical Developer
Joined
Apr 25, 2019
Messages
711
I always wondered why software (MyBB has it too) liked implementing threaded mode, a feature with next to no practical utility, it seems it was implemented to tick boxes, now that isn't a rub on you in particular, a lot of software make the mistake of chasing features rather than methodically trying to weed out features.

It is easy enough adding one feature, but when you add and add and add, then before long, you have a big scary chimera on your hands. You have to wade through dozens of features which don't quite mesh well, but which someone thought "might be nice" to get to the good stuff.

There was a good article on this somewhere about how far down the rabbit-hole "just one more feature" can take you, but I can't find it, needless to say it was a chat which ended up with so many auxiliary features that users are absolutely overwhelmed.
 

LeadCrow

Apocalypse Admin
Joined
Jun 29, 2008
Messages
6,562
I always wondered why software (MyBB has it too) liked implementing threaded mode
It's a legacy feature that gained a massive resurgence about a decade ago but not on forums. vbseo's like system granted visibility to posts based on their like counts in addition to chronological ordering, which makes quality replies surface. At the time, upvote systems were still considered vilified on forums, wether it was for rating members or rating posts and their display often hidden.
Digg then reddit popularized threading mode big time as they made ordering dependant on 'upvotes' rather than chronological.

Forums should've reconsidered the merits of doing only chronological and reverse-chronological ordering among others. Neither suits long discussions particularly well and both dilute the value of your users' replies instead of having the best content surface no matter how long ago it was posted. Maybe the forum industry should pool together to discuss ways to improve the entire ecosystem instead of everyone shunning collaboration and hoping for the demise of every other script while social media's and platforms are burning the entire platform they stand on.
 

cheat_master30

Moderator
Joined
Jan 16, 2010
Messages
3,855
I always wondered why software (MyBB has it too) liked implementing threaded mode, a feature with next to no practical utility, it seems it was implemented to tick boxes, now that isn't a rub on you in particular, a lot of software make the mistake of chasing features rather than methodically trying to weed out features.

It is easy enough adding one feature, but when you add and add and add, then before long, you have a big scary chimera on your hands. You have to wade through dozens of features which don't quite mesh well, but which someone thought "might be nice" to get to the good stuff.

There was a good article on this somewhere about how far down the rabbit-hole "just one more feature" can take you, but I can't find it, needless to say it was a chat which ended up with so many auxiliary features that users are absolutely overwhelmed.
My thought was that MyBB was always inspired by vBulletin, and vBulletin 3 also had a threaded mode aroud the same time.

As for why they were added? Well, because some older forum communities originally used threaded discussions, and these scripts wanted to appeal to them.
 
Top