DEV Community

Cover image for Stop rewarding quantity!
Ingo Steinke
Ingo Steinke

Posted on

Stop rewarding quantity!

I have received a DEV badge to honor consistency of writing, as I had published at least one post per week for eight weeks. Notifying me of the badge, DEV wants to encourage me to continue the streak and receive another badge for posting something (anything, actually) every week for another eight weeks. I could even write a bot who does that, and then I'll write another post featuring a codepen of my bot script.

What's the hype about consistency? πŸͺ₯

After a thoughtful second, I decided to put the badge on my not-to-do-list.

While building habits can help people to achieve their goals, I think that rewarding quantity is a step in the wrong direction (unless it's about brushing your teeth every day). Our society is already based on routine, making people drive to their 9-to-5-jobs in their cars day by day, so it feels like a genius "life hack" to replace some of our old routines by other, slightly less toxic ones, and "earning" imaginary badges in our habit tracking apps or gratefulness journals.

I could post a rant every day ...

For the sake of consistency of writing, I could also post a little rant about any annoying misfeature every day. I don't mean DEV misfeatures, but rather anything digital, Android, Apple, Google, Microsoft, JavaScript, or some bad idea a customer absolutely thinks I must develop that does not have proper native support in browsers. It won't be hard to publish something ungrateful every day.

... or lower my standards and don't finish my work

Last, but not least, I could make less of an effort when researching and learning new features and make sure to release a new post every week no matter if I'm finished with what I set out to achieve. Just let the community correct my mistakes and fill in the blanks. DEV would reward the achieved consistency with another badge in eight weeks from now, despite the decreased quality of my content.

I can also split up my content in much smaller pieces to publish series, so instead of my conclusion, you'd get a cliff hanger quote to meditate about for another week.
Sometimes that makes sense, but often, it does not.

Badges are pretty pointless πŸ†

As INHU pointed out, "badges are pretty pointless". Which might seem funny to hear from someone who set their own goals accouncing a huge quantity of new post in 2022.

Quantity allows no conclusion about quality. A continuously growing pile of bullshit is still a pile of bullshit. No personal offense: some people release a lot of content while providing high quality at the same time, and @inhuofficial is probably on of those people.

@thormeier also seems to find it easy enough to deliver weekly content and use the writing process to push themselves beyond their comfort zone:

But while quality and quantity are unrelated in theory, often we have to decide to prioritize either one of them.

Side effects of gamification πŸ’©πŸ’©πŸ’©

Gamification is a controversial approach to reward desired behavior, risking to boost unwanted side effects instead. StackOverflow has been trying to build a "meritocracy" focused on high-quality technological discussions, but they are still pretty far from reaching their goal in my opinion, often giving the impression of a gatekeeping community proud of their museum of outdated jQuery code snippets, but that doesn't mean it was wrong to try.

Irrelevant clickbait and dangerous advice

DEV, as a very open-minded and welcoming online community, has proved a source of information, inspiration, discussion, and entertainment. But, besides many articles being irrelevant to me personally (while still having value for other readers), DEV also attracts a lot of dubious content with spammy headlines like "22 node modules / react plugins EVERY developer MUST know in 2022" featuring copy-and-pasted half-knowledge, sometimes even giving risky advice contrary to established best-practice consensus.

Is DEV a victim of its own success?

@samuelfaure, who left Medium "after it became shit", elaborated on the problems that arise from popularity.

As DEV allows everyone to write to the best of their ability, which is a good thing, it is still a great benefit for the community to reward and promote outstanding quality content.

Promoting outstanding quality content πŸ™

DEV already has topical hashtags and options to indicate the difficulty level, and they have a team that reads, awards and curates content. I can't thank them enough! Their helpfpul expertise and effort could never be achieved by an algorithm.

Using hashtags and experience level as filters

Hashtags like #beginners or #javascript could be used to filter by personal preference and interests and help which content to feature on certain landing pages automatically. Maybe we could use an option to suggest changing hashtags or difficulty classifications of articles that we have read and find they are not tagged accordingly.

But there is one simple change that might make much of a difference:

Stop rewarding quantity!

Badges are nice, especially if someone wrote an article that got hand-picked into a best of the week or best topical write-up curation.

But why would we want to reward consistency? I like to create high quality content, include working demo code and make sure my claims are factually correct and proved by authoritative citations. That benefits my readers, and it benefits my own learning process at the same time. That takes time to do properly, and we should rather take our time and release one good article per month instead of feeling obliged to deliver something every week.

At least, that's my own opinion, also due to the fact that I see a growing number of articles on DEV that might have better been saved as a draft instead of being published in their current state.

Collage of the 8-weak-streak badge, and emoji clip art graphic of a calendar with weekly checkmarks, a diamond, and a lot of poo pile characters

Being tempted to release a half-baked post just to get on with a continuity streak, or release additional "inspirational" posts with calendar motto bullshit does not help to improve quality at all. DEV is neither StackOverflow nor Wikipedia, but it doesn't have to become the next LinkedIn or Medium either. There are already too many platforms promoting mediocre bulk content.

Prioritize quality over quantity

Let's take our time to finish well-written articles and don't let the hollow goal of a continuous streak divert us from prioritizing quality over quantity!

Discussion (18)

Collapse
inhuofficial profile image
InHuOfficial • Edited on

Thanks for the shout out bud.

I think perhaps I should state that the consistency badge is one of the few that I actually like!

Yes it could encourage half-baked articles, you are correct (in fact I think it did result in me doing a "quick post" one week to continue the streak, which certainly adds validity to what you are saying!).

However consistency in writing is a massive win if you are new to writing and even if one of my contributions was not great it did encourage me to write when I might not have felt like it and subsequently enjoy writing the article!

As with anything it is a double edged sword!

As one of the people who loves DEV but will also criticise anything I think they are not doing correctly, I would also say that things seem to be improving, you do get the odd days where the feed looks like something from "the spam societies guide to spamming" but overall I think it is certainly on the right trajectory (the feed feels better now than 6 months ago and fewer "10 VS Code plugins that you must use" type posts seem to get into the top of the week position etc.).

I think the key to a good experience on DEV is to follow the people who write good content, whenever I see a notification from you or @afif, @alvaromontoro, @thormeier, @dailydevtips1 etc. I know I am going to enjoy it. (this is far from an exhaustive list of people who's articles I like!)

I do not envy the DEV team, trying to make a feed that rewards quality over pandering listicles is very difficult!

For example you said about the tags system, but that gets abused (add #javascript, #beginners and guarantee to get views!), manual moderation is linked to company biases so that becomes difficult to get right (if everyone in the company loves Rust for example then more Rust articles will fill the feed if manual intervention takes place) etc.

There are no great solutions, only better ones as every system can be gamed and is a victim of "give the people what they want to see", and a lot of people want to see listicles and click on clickbait titles (as you know I always walk a fine line where I make something appear clickbait but then (hopefully) deliver!).

It saddens me that authors like @samuelfaure don't like the platform anymore, I do understand, but at the same time I think all of the alternatives have similar issues and at least the DEV team appear to care a lot (even when I "butt heads" with them they are still passionate about their product and seem to care about quality)!

I think the key might be something I was thinking about for a while, which is to hand pick authors to form a "power team" of super high quality content and put them all together on a site. I own a2z.dev which would be an awesome domain for varied writers, so who knows, maybe we just all band together!

Anyway, I have gone off track, lots of food for thought, I will round up by saying it is a thought provoking and well written article and thank you! β€πŸ¦„

Collapse
ingosteinke profile image
Ingo Steinke Author

Tags are more flexible than categories or topics of classic bulletin board discussion forums like daniweb, reddit, or quora, none of which I like to use. Probably I wouldn't really like Usenet (aka newsgroups) anymore as well. DEV offers discussion and inspiration without the visual ugliness of bulletin boards but still on a higher technical level than medium, tealfeed etc.

The hand picked authors idea seems appealing, but how are those biased who are supposed to pick? This might lead to a similar "always the same" feeling that some conferences already suffer, and once we start to focus on the power team blog, who would still follow and engage with newcomers on DEV to recognize when some of the listicle noobs have evolved to become authors we might want to feature?

Collapse
inhuofficial profile image
InHuOfficial

To get β€œfresh blood” into the group I would suggest that as the authors stumble across great articles they explore someone’s other work. If they consistently produce high quality articles (I mean majority are high quality not so much how often) then the group votes on whether they get an invite. Sounds a bit elitist when I say it like that but I hope you get my gist lol!

Collapse
samuelfaure profile image
Samuel FAURE • Edited on

Thanks for continuing on this discussion.

Let's hope Dev.to won't keep going that way. I personally never hang out here anymore. The signal/noise ratio is so bad I'd have better chances to get interesting dev insight from a broken TV.

Collapse
ingosteinke profile image
Ingo Steinke Author • Edited on

Oh dear! I totally agree about the signal/noise ratio, but I think it would be unfair to ditch DEV altogether, as there is still a lot of good content (even if you're not an aspiring React developer).

I tend to ignore any kind of activity feed, be it chronological or algorithmic. Just have a look at f***book and twitter. Feeds like that might be valuable for moderators and curators, but otherwise we need search, tags or categories, and manual curation and linking.

I hope I made my point that much of the "noise" from an experienced perspective can be full of valuable insights for beginners, but I doubt that all of them are able to distinguish the valueable ones from the bullshit, and that trying to weed out low-quality content the StackOverflow way is not a good alternative either.

Collapse
ingosteinke profile image
Ingo Steinke Author
Collapse
valeriavg profile image
Valeria

My feed is particularly shitty today, may be due to the ongoing A/B testing or the position of stars, but still...

It's discouraging and it seems that I'll need to "optimise" my articles if I expect them to be read.

So, after careful consideration, I decided to stop giving a crap and join this rant instead 😁

The only reason I used this platform for so long is because I didn't need to care about anything but actual writing. Not consistency, not hyped topics, just writing articles. The better they were - the better they performed.

But lately even the weekly roundup promotes some fairly mundane creations, so I'm clearly not the target audience here either.

I hope it's the A/B test, though, I really do...

Collapse
ingosteinke profile image
Ingo Steinke Author

I also tried to A/B test on optimizing articles, but not in a systematic way, and with so little articles and so many other factors (like @inhuofficial observed some "ideal times to release") I doubt it's anywhere above randomness either.

Probably there are some ways to optimize articles that don't require much effort and don't change your style of writing.

I tried to use emojis in headings and even in article titles, I tried giving more or less information in the title, adding more or less images, code, codepens, and third party content like videos, and I think this is the first article where I used the proper liquid tags to embed articles ({% post ) and mention other authors (using the at sign).

But like many others already said that their most important audience is the author, that they are writing for themselves, and if anybody else is interested, even better.

Collapse
valeriavg profile image
Valeria

Yup, I gave it it a try with "Learn X in just Y minutes", which got bunch of reactions, but today I've got a comment on my old "non-optimimal" article. User was thanking me for it and showing how they've used the technique I shared in their project. I'm still grinning, to be honest. I'd trade a thousand likes for such comment any day.

Collapse
dailydevtips1 profile image
Chris Bongers

I agree with @inhuofficial here, it's definitely a double-edge sword.

I don't even know what badges I have, and I don't care.
I write consistency for me, not for you, your neighbour, or anyone else.

It's for me.

And if these badges can help beginners form a habit, i'm all for it.
I think the hyped listicles happen to any form of person, not just people who need to push out articles, but those who think they will get better from it.

I've noted a lot of people wanting to get into technical writing, because they think it will make them rich.

Either i'm doing something wrong, of their believes are off, because I'm not rich πŸ˜‚

Anyway, point is, it will always happen, because it works!
Look at the top articles "ever" on dev.to they are 9/10 listicles.

So where lies the real problem?
I think in people clicking and engaging in those articles.

Don't ask me why, because I certainly don't enjoy them (and yes I have written some in the past shame).
However there is a matter where I think they can even work fine, if executed correctly and bringing value to a point.

Alright ending this comment with:
Don't choose quantity over quality!

Consistency rocks, but choose something that works for you

Collapse
thormeier profile image
Pascal Thormeier

Thank you for the shout out! Glad to see that people care about quality and start constructive discussions.

If you open up a platform for everybody, you will get all kinds of levels of quality. I always thought that the curated "Top 7 of the week" and the top posts per tag do offer some measurement of quality, though. Good content will be shared, thus more engagement, thus more reads, reactions and comments, right? Even if it was just the millionth To-do list tutorial for the day, if it was an exceptionally good one, why not reward that? Gamification works, it just needs to be insanely fine-tuned in order to achieve what the platform wants - high amount of high quality content.

What about this: You do get streak badges (as in 8 weeks, 16 weeks, etc.), but they're tied to the articles being of good quality? For example, 4 "Top 7 of the week" posts in 4 weeks, or 8 top tag badges in 8 weeks? It would put high pressure on the quality level and make those streak badges a lot harder to achieve, encouraging consistency and quality?

Collapse
ingosteinke profile image
Ingo Steinke Author

Thanks for your comment! A combination of consistency in time plus high quality reminds me of some StackOverflow badges (like the "curious" ones I never got, as most of my questions got deleted as offtopic or attracting opionionated answers). Might be a good idea, while nothing for the mass and the aspiring newcomers.

But "top of the week" etc. still seems the best way to measure some sort of quality at least. Maybe they should consider reading time as well, so that just clicking on a spammy headline without at least scrolling halfway through the article content would not count as a view.

Collapse
flodev profile image
Florian Klenk • Edited on

Hi, thanks for making a point.
The same principle should be applied for delivering source code. The less code the better. Dev.to should actually be aware of it... but apparently that's not the case πŸ˜•
But to be fair quantity is much easier detectable than quality πŸ˜†

Collapse
ingosteinke profile image
Ingo Steinke Author • Edited on

good point Florian! I heard that developers used to get paid by lines of code in early days of software development. Looking for a source, I found this (not finished reading though)

The oldest metric for software projects is that of β€œlines of code” (LOC). This metric was first introduced circa 1960 and was used for economic, productivity, and quality studies

ifpug.org/content/documents/Jones-...

Collapse
ingosteinke profile image
Ingo Steinke Author

and getting paid by hours spent in an office or in front of the monitor by default these days is only slightly better, at least it's no longer about output quantitaty

Thread Thread
flodev profile image
Florian Klenk

This is what I really hated in most companies. Being there just for being there 9 to 5. Everyone especially the management distrusts each other to just screw around when they have no tasks.
In my opinion tasks should be planned for a sprint and when the tasks are done it's done. Every employee can use the time that's left to learn but does not have to.
There has to be some slack so that employees can be creative.

Collapse
afif profile image
Temani Afif

don't forget to add tags to your post to give it some visibility

Collapse
moopet profile image
Ben Sinclair

As far as gamification goes, I believe I'm better off without it. I believe most people are, too, so I use a bunch of plugins, like Twitter Demetricator for example, to remove the like/retweet counts from Twitter.

Since I can't control what appears on my feed (that's down to other people falling for the popularity contest) it doesn't make a huge amount of difference to what I read, but I don't want numbers to make a difference to what expectations I have of an author or whether I'm inclined to vote for them one way or another.

I'm definitely not a fan of badges. I like unlocking different privileges on Stack Overflow as time goes by becuase they're related to what I've done and the process is generally so slow that by the time I can, say, re-open a ticket, I've been there and seen enough arguments about it to know what I'm doing.

Without going to my profile, I have no idea what badges I have, except a couple of "1 year!" anniversary badges which are just an accident of when I signed up. It doesn't matter. I'm not going to look at someone else's profile and be impressed, neither am I going to decide not to follow them because they don't have enough bling.