On Tuesday, 2 October 2018 at 06:26:30 UTC, Joakim wrote:
>[snip]

Also you're out by a year :)

On Tuesday, 2 October 2018 at 09:39:14 UTC, Adam Wilson wrote:
> On 10/1/18 11:26 PM, Joakim wrote:
>> [snip]
>
> I disagree.

It is not clear what you disagree with, since almost nothing you say has any bearing on my original post. To summarize, I suggest changing the currently talk-driven DConf format to either

1. a more decentralized collection of meetups all over the world, where most of the talks are pre-recorded, and the focus is more on introducing new users to the language or

2. at least ditching most of the talks at a DConf still held at a central location, maybe keeping only a couple panel discussions that benefit from an audience to ask questions, and spending most of the time like the hackathon at the last DConf, ie actually meeting in person.

Since both of these alternatives I suggest are much more about in-person interaction, which is what you defend, and the only big change I propose is ditching the passive in-person talks, which you do not write a single word in your long post defending, I'm scratching my head about what you got out of my original post.

> There is much more to the conference than just a 4-day meetup with talks. The idea that it's just the core 8-15 people with a bunch of hangers-on is patently false. It's not about the conversations I have with the "core" people. It's Schveighoffer, or Atila, or Jonathan, or any of a long list of people who are interested enough in coming. Remember these people self-selected to invest non-trivial treasure to be there, they  are ALL worthy of conversing with.

Since both my mooted alternatives give _much more_ opportunity for such interaction, I'm again scratching my head at your reaction.

> Is it a "mini-vaction"? Yea, sure, for my wife. For her it's a four day shopping spree in Europe. For me it's four days of wall-to-wall action that leaves me drop-dead exhausted at the end of the day.

So it's the talks that provide this or the in-person interaction? If the latter, why are you arguing against my pushing for more of it and ditching the in-person talks?

> Every time I see somebody predicting the end of "X" I roll my eyes. I have a vivid memory of the rise of Skype and videoconferencing in the early 2000's giving way to breathless media reports about how said tools would kill the airlines because people could just meet online for a trivial fraction of the price.

People make stupid predictions all the time. Ignoring all such "end of" predictions because many predict badly would be like ignoring all new programming languages because 99% are bad. That means you'd never look at D.

And yes, some came true: almost nobody programs minicomputers or buys standalone mp3 players like the iPod anymore, compared to how many used to at their peak.

> However, it's 2018 and the airlines are reaping record profits on the backs of business travelers (ask me how I know). Airlines are even now flying planes with NO standard economy seats for routes that cater specifically to business travelers (e.g. Singapore Airlines A350-900ULR). The order books (and stock prices) of both Airbus and Boeing are at historic highs.

You know what is much higher? Business communication through email, video-conferencing, online source control, etc. that completely replaced old ways of doing things like business travel or sending physical packages. However, business travel might still be up- I don't know as I haven't seen the stats, and you provide nothing other than anecdotes- because all that virtual communication might have enabled much more collaboration and trade that also grew business travel somewhat.

> There are more conferences, attendees, and business travelers than there has ever been in history, in spite of the great technological leaps in videoconferencing technology in the past two decades.
>
> The market has spoken. Reports of the death of business/conference travel have been greatly exaggerated.

You are conflating two completely different markets here, business versus conference travel. Regarding conferences, your experience contradicts that of the iOS devs in the post I linked and the one he links as evidence, where that blogger notes several conferences that have shut down. In your field, it is my understanding that MS has been paring back and consolidating their conferences too, though I don't follow MS almost at all.

> The reason for this is fundamental to human psychology and, as such, is unlikely to change in the future. Humans are social animals, and no matter how hard we have tried, nothing has been able to replace the face-to-face meeting for getting things done. Be it the conversations we have over beers after the talks, or the epic number of PR's that come out the hackathon, or even mobbing the speaker after a talk.

It is funny that you say this on a forum where we're communicating despite never having met "face-to-face," discussing a language where 99.999% of the work is done online by people who don't need any "face-to-face" meetings to get "things done." :)

Also, my suggestions are about enabling more face-to-face time, not less, so there's that too.

> Additionally, the conference serves other "soft" purposes. Specifically, marketing and education. The conference provides legitimacy to DLang and the Foundation both by it's mere existence and as a venue for companies using DLang to share their support (via sponsorships) or announce their products (as seen by the Weka.io announcement at DConf 2018) which further enhances the marketing of both the product being launched and DLang itself.

Don't make me laugh: what part of this marketing/legitimization couldn't be done at either of the two alternatives I gave?

> I have spoken to Walter about DConf numerous times. He has nothing against, and indeed actively encourages, local meetups. But they do not serve the purpose that DConf does. My understanding from my conversations with Walter is that the primary purpose of DConf is to provide a venue that is open to anyone interested to come together and discuss all things D. He specifically does not want something that is only limited to the "core" members. As this suggestion runs precisely counter to the primary stated purpose of DConf it is unlikely to gain significant traction from the D-BDFL.

Wrong, both of my suggestions fulfil that purpose _better_. What they don't do is limit attendance to those who have the passion _and_ can afford the time and money to travel 2-20 hours away to a single location, just so they can get all the in-person benefits you claim.

> Yes, it is expensive, but in all the years I've attended, I have not once regretted spending the money. And indeed, coming from the west coast of the US, I have one of the more expensive (and physically taxing) trips to make. I know a number of people who found jobs in D through DConf, would that not make the conference worth it to them?

How many people got jobs versus how many attended? Would that money to get 100 people in the same room seven times have been much better spent on other things?  Run the cost-benefit analysis and I think it's obvious my two suggestions come out better. At best, you can maybe say that wasn't the case at the first DConf in 2007, when high-speed internet wasn't as pervasive and Youtube was only two years old, but not for every DConf since.

> Something is only expensive if you derive less value from it than it costs. And for many people here, I understand if the cost-benefit analysis does not favor DConf. But calling for an end to DConf simply because it doesn't meet someones cost-benefit ratio is inconsiderate to the rest of us who do find the benefit.

I don't care about your personal cost-benefit ratio. I care about the cost-benefit analysis to the language and ecosystem as a whole.

> Nobody is making you go, and, since you already get everything you want from the YouTube video uploads during the conference, why do you care if the rest of us "waste" our money on attending the conference? That is our choice. Not yours.

Try reading the older forum thread I originally linked, Jonathan and I have already been over all this. D is a collective effort, and it's a colossal waste of the community's efforts to spend all that time and money on the dying conference format that DConf has been using.

It signals to me and many others that D is not a serious effort to get used as a language, but simply a bunch of hobbyists who want to have "fun" meeting up at an exotic locale once a year, in between hacking on an experimental language that they're fine if nobody else uses.

If that's D's focus, fine, just own it. Put it on the front page: "This is a hobbyist language, please don't bother using it in production. We are much more focused on where we can vacation together next year than trying to spread awareness and improve the language."

Regardless of whether you post that notice or not, that is what continuing the current DConf format advertises, given that others have already been moving away from it.

> Note: Limiting anything to "core" members is a guaranteed way to create a mono-culture and would inevitably lead to the stagnation of D.

Good, then you agree with me that we should avoid such stagnation by broadening DConf to be a bunch of meetups in many more cities?

> Which is why anybody can post to all NG's, even the internals NG.

This is not actually true. There are two newsgroups that seem to have that designation, which show up separately as `internals` and `dmd` at forum.dlang.org, and the latter doesn't allow me to post to it without registering somewhere, unlike the rest of the web forums.

Guess what the current DConf format does to most people who don't attend too...

I'm done responding to these irrational responses that ignore everything I wrote. I'll just link them to this long debunking from now on.

On Tuesday, 2 October 2018 at 10:37:44 UTC, Nicholas Wilson wrote:
> On Tuesday, 2 October 2018 at 06:26:30 UTC, Joakim wrote:
>> [...]
>
> As I'm sure has been said before, if it were just the talks it probably wouldn't be worth it. But conferences are sooooooooooo much more than just the talks. Its the conversations over breakfast/lunch/dinner/ between talks and long into the night (sometimes too long). Its the networking, the hacking, the face to face. The talks are usually pretty good too.
>
> The conference is definitely not dead, I'm going to one in San José in 2 weeks, sure the talks look really interesting but the main reason is to talk to other people to get stuff done.

Then I'm not sure why you're saying any of this to me, as almost nothing you write contradicts anything I wrote.

If you're still not sure what I mean, read this long post I just wrote fisking Adam's similar post:

https://forum.dlang.org/post/eoygemytghynpogvljwb@forum.dlang.org

[snip]

However constructive your alternate proposals are, I suspect people are misreading your title (and it is easy to assume, just from the OP title, that you actually want to get 'get rid of' DConf, rather than just 'modify' and 'improve' DConf.

Personally I feel there is a cognitive dissonance between your title and what you actually wrote. Maybe, since the people who so far responded seem really vested in the benefits of DConf.. there might be a bit of knee jerking going on here.

On Tuesday, 2 October 2018 at 06:26:30 UTC, Joakim wrote:

> "Once the videos are all up, set up weekend meetups in several cities [all over the world], where a few livestreamed talks may talk place if some speakers don't want to spend more time producing a pre-recorded talk, but most time is spent like the hackathon, discussing various existing issues from bugzilla in smaller groups or brainstorming ideas, designs, and libraries for the future."
>
> I can setup an event like this in my city, where AFAIK nobody uses D, so most of it would be geared towards introducing them to the language.
>
> I estimate that you could do ten times better at raising awareness and uptake with this approach than the current DConf format, by casting a much wider net, and it would cost about 10X less, ie you get two orders of magnitude better bang for the buck.

I think this is something that could be done *in addition to* DConf. I honestly don't think DConf is very effective at promoting D, except perhaps to a small sliver of the overall population of programmers, due to the content of most of the presentations. {This is not intended to be a criticism or a statement that anything about DConf should be changed.}

I believe it would be a mistake to drop DConf. If we did that, the story that would be told is "D couldn't even support its own conference. Use Rust or Go or Julia instead." Our view would be "we're on the cutting edge" but everyone else's view would be "the language is dying".

On Tuesday, 2 October 2018 at 14:49:31 UTC, bachmeier wrote:
> I believe it would be a mistake to drop DConf.

What about we design a DConf that focuses on interactive collaboration instead of sitting passively in a room watching someone talk over a slideshow?



When Joakim talked about this the last time, I was just getting home from a work trip that got 30 people from the company - who usually work online - together in person. We had very few talks in the style of old DConf, and instead we would do a group introduction thing, then usually break off into smaller (randomized) work groups to progress a solution to the problem presented in the talk, then get back together as a whole to discuss it. (Etc.; there were a few different styles we tried, but most of them basically followed this basic idea.)

BTW another thing we did is the whole agenda was set up ahead of time, and there were some pre-reading we could do on the flights there, so we arrive already familiar with the ideas and might have some thoughts in mind already. This is analogous to putting the talks on Youtube first, then talking about it/working on it in person.


That also happens to be what most people say they like most about DConf... but at DConf, it happens after-hours, since the main event is all one talker presenting a powerpoint while everyone else <strike>surfs the internet</strike> listens intently. At my work meeting, the whole thing was designed around this in-person interaction.

That is what Joakim is talking about - changing the main event to be more like the after-hours stuff everyone loves so much, to actually use all the time to maximize the potential of in-person time.

On Tuesday, 2 October 2018 at 14:49:31 UTC, bachmeier wrote:
> On Tuesday, 2 October 2018 at 06:26:30 UTC, Joakim wrote:
>
>> "Once the videos are all up, set up weekend meetups in several cities [all over the world], where a few livestreamed talks may talk place if some speakers don't want to spend more time producing a pre-recorded talk, but most time is spent like the hackathon, discussing various existing issues from bugzilla in smaller groups or brainstorming ideas, designs, and libraries for the future."
>>
>> I can setup an event like this in my city, where AFAIK nobody uses D, so most of it would be geared towards introducing them to the language.
>>
>> I estimate that you could do ten times better at raising awareness and uptake with this approach than the current DConf format, by casting a much wider net, and it would cost about 10X less, ie you get two orders of magnitude better bang for the buck.
>
> I think this is something that could be done *in addition to* DConf.

It depends what you mean by that. If DConf keeps running as it has, as you suggest below, but you simply add some satellite meetups around it in other cities watching the livestreamed talks from the main DConf, then you have addressed some of these concerns, but not very much.

If you go the decentralized approach I suggested, but maybe pick one of those locations as the one the core team goes to and don't do almost any in-person talks anywhere, that would address much more.

> I honestly don't think DConf is very effective at promoting D, except perhaps to a small sliver of the overall population of programmers, due to the content of most of the presentations.

I agree. I'll go farther and say that it's a small sliver of existing D programmers too who get much value out of it.

> {This is not intended to be a criticism or a statement that anything about DConf should be changed.}

Heh, of course it's a criticism and of course it should be changed. :)

> I believe it would be a mistake to drop DConf. If we did that, the story that would be told is "D couldn't even support its own conference. Use Rust or Go or Julia instead." Our view would be "we're on the cutting edge" but everyone else's view would be "the language is dying".

Great. Everybody thought Apple was nuts when they released a $500 iPhone in 2007, now Ballmer wishes he'd come up with the idea:

https://www.macrumors.com/2016/11/07/former-microsoft-ceo-steve-ballmer-wrong-iphone/

As long as you communicate that you're replacing one DConf location with several and why you're doing it, I don't see why we should care how they end up interpreting it. Our goal is to get users and adoption, not to look good to other programming-language developers.

On Tuesday, 2 October 2018 at 15:03:45 UTC, Adam D. Ruppe wrote:
> That is what Joakim is talking about - changing the main event to be more like the after-hours stuff everyone loves so much, to actually use all the time to maximize the potential of in-person time.

I'm talking about growing two different qualities much more, with my two suggested alternatives to the DConf format.

1. Ditch the talks, focus on in-person interaction. That's why I suggest having almost no talks, whether at a central DConf or not. You clearly agree with this.

2. Decentralize the DConf location, casting a much wider net over many more cities. Walter and Adam could rent a room and setup a Seattle DConf location, Andrei and Steven in Boston, Ali and Shammah in the bay area, and so on (only illustrative, I'm not imposing this on any of these people). Some of the money that went to renting out a large conference room in Munich can instead be spent on these much smaller rooms in each city.

Charge some minimal fee for entrance in some locations, if that means they can spend time with W&A and to cover costs. I wouldn't charge anything more than $2 in my city for my event, as event organizers here have found that that's low enough to keep anyone who's really interested while discouraging fake RSVPs, ie those who have no intent of ever showing up but strangely sign up anyway (I know an organizer who says he had 150 people RSVP for a Meetup here and only 15 showed up).

By keeping travel and ticket costs much lower, you invite much more participation.

Obviously my second alternative to DConf listed above wouldn't be decentralized at all, only enabling in-person interaction at a still-central DConf.

Mix and match as you see fit.

On Tuesday, 2 October 2018 at 15:42:20 UTC, Joakim wrote:
> On Tuesday, 2 October 2018 at 15:03:45 UTC, Adam D. Ruppe wrote:
>> That is what Joakim is talking about - changing the main event to be more like the after-hours stuff everyone loves so much, to actually use all the time to maximize the potential of in-person time.
>
> I'm talking about growing two different qualities much more, with my two suggested alternatives to the DConf format.
>
> 1. Ditch the talks, focus on in-person interaction. That's why I suggest having almost no talks, whether at a central DConf or not. You clearly agree with this.
>
> 2. Decentralize the DConf location, casting a much wider net over many more cities. Walter and Adam could rent a room and setup a Seattle DConf location, Andrei and Steven in Boston, Ali and Shammah in the bay area, and so on (only illustrative, I'm not imposing this on any of these people). Some of the money that went to renting out a large conference room in Munich can instead be spent on these much smaller rooms in each city.
>
> Charge some minimal fee for entrance in some locations, if that means they can spend time with W&A and to cover costs. I wouldn't charge anything more than $2 in my city for my event, as event organizers here have found that that's low enough to keep anyone who's really interested while discouraging fake RSVPs, ie those who have no intent of ever showing up but strangely sign up anyway (I know an organizer who says he had 150 people RSVP for a Meetup here and only 15 showed up).
>
> By keeping travel and ticket costs much lower, you invite much more participation.
>
> Obviously my second alternative to DConf listed above wouldn't be decentralized at all, only enabling in-person interaction at a still-central DConf.
>
> Mix and match as you see fit.

I totally agree with you on your first point, i.e. making DConf more interactive. I have had very good experiences with formats like open space or barcamp. However, these formats only work if people are actually willing to participate and bring in their own ideas. Not having anything prepared can in rare cases lead to the situation where there is a lack of things to talk about (I doubt this would be the case for the D community, but it is something to keep in mind).

However, I must say I disagree with your second point, i.e. decentralising DConf. As many people here have already mentioned, DConf is about talking to people. And to me it is especially important to talk to lots of different people whom I otherwise don’t get the chance to talk to in person. By decentralising the conference, we would limit the number of different people you can get in touch with directly by a huge amount.

Just to use myself as an example, last Docnf I was able to talk to Andrei, Walter, Mike, Ali, Jonathan, Kai and lots of others and exchange ideas with them. This would not have been possible with a decentralised event (except for the off chance that all those people by chance attend the same local „meetup“).

On the other hand, I have to admit that decentralising the event would open it up for a much bigger audience, which definitely is a good idea. However, I would much prefer to have something like a main DConf and if there are enough interested people in an area who will not go to the main event, they can host their own mini conference and watch streams, make their own small workshops etc. This is what happens a lot at the Chaos Communication Congress and it seems to work really well (granted, in this case it might also be related to the limited number of tickets).

On Tuesday, 2 October 2018 at 16:10:20 UTC, Johannes Loher wrote:
>
> Just to use myself as an example, last Docnf I was able to talk to Andrei, Walter, Mike, Ali, Jonathan, Kai and lots of others and exchange ideas with them. This would not have been possible with a decentralised event (except for the off chance that all those people by chance attend the same local „meetup“).
>

And local decentralized meetups already existing. In fact I'm co-organizing one here in Denmark.

Similarily there are a lot other local meetups other places around the world.

DConf is a great way to centralize all those meetups so you get to meet people from different cultures with different views on things, because whether you're aware of it or not then programming is done different in every country, because each country has their own technological needs, culture etc. and it shapes very much around that.

Ex. a banking system in America is not the same as a banking system in Germany.