Great to hear of small successes in reception of Pharo in the broader community. Some ideas...

1. The most kick ass thing I can think of would go something like this... a. Set up two computers nominally "customer" (or "webserver") and "support team". Start the same image on each. b. On the "customer" computer run some code that cause an error. In the debugger Fuel-out to a USB stick. Show them the small size of the file. c. On the "support team" computer, insert the USB stick and drag the Fuel file into the image and start debugging.

Or the CI example were test failures are fueled out. This is a nice extension to the typical CI behavior.

Even better if the code that was running was a morph stepping across the screen with some text inside counting up, and that counter and morph position were maintained in the transition to the second computer - but I'm not sure if Fuel would handle that.

There is no way to know it but trying. The only problem could be that during the serialization, the traversed graph to serialize may be to big if it reaches to the display and all its world and related objects. Also, you may arrive to places that change so frequently like mouse position etc... This normally happens when you end up serializing the whole UI. In this case, it is likely the graph will change while you are serializing it...so you will be screw. Of course, if can hook a bit and cut the graph...but you should know in which place of your morph to cut.

2. Parts of the Zinc quick start might be worthwhile... http://zn.stfx.eu/zn/build-and-deploy-1st-webapp/

3. Pull up halos on a menu and drag a duplicate a menu item onto the background, then show that it still works when it is clicked. Bring up the halos again and inspect the morph where you can change the name of that menu item. Exit Pharo and restart to show that the renamed menu item is still there.

4. Some of Doru's "build a custom browser" demos of Glamorous Toolkit are inspiring for how few lines of code are needed, but I don't have references.

good luck, cheers -ben

-- Mariano http://marianopeck.wordpress.com

how about the other way around ? Pharo being the front end , the IDE for amber, using Pharo to live code a website. That would be a very cool demo. Especially if Amber could leverage the Pharo debugger. On Tue, Apr 29, 2014 at 11:58 AM, Markus Fritsche <mfritsche@reauktion.de>wrote:

On 2014-04-28 17:17, Ben Coman wrote:

1. The most kick ass thing I can think of would go something like this...

The most "kick ass thing" I can think of is...

- Have a Pharo World being interfaced to an amber frontend, so you can use your image through your browser.

... on second thought... that might be cool but after five minutes it'll make me think "and now?"

;-)

Esteban Lorenzano writes:

...

Pharo being the front end , the IDE for amber, using Pharo to live code a website. That would be a very cool demo. Especially if Amber could leverage the Pharo debugger.

I’m trying to convince Nico to do that for *ages* but he does not sees the advantages/coolness of that approach.

Esteban, That is just not true (and unfair). I in fact started a project called Nemo with Ben some time ago but my time is limited just like everyone else. People have worked on it since then, but AFAIK it's still far from being complete. Cheers, Nico

On 29 Apr 2014, at 15:42, Sebastian Sastre <sebastian@flowingconcept.com> wrote:

On Apr 29, 2014, at 10:22 AM, Nicolas Petton <petton.nicolas@gmail.com> wrote:

...

...

I’m trying to convince Nico to do that for *ages* but he does not sees the advantages/coolness of that approach.

Esteban,

That is just not true (and unfair).

I in fact started a project called Nemo with Ben some time ago but my time is limited just like everyone else. People have worked on it since then, but AFAIK it's still far from being complete.

My bad. I’m sorry.

...

Cheers, Nico

sounds cool

But we should address that problem, right?

The only way I see is by throwing more talent on the problem

Here is one thing that accelerates that process and catalyses everything and we are NOT doing:

1. workshops, barcamps, talks, breaking our faces shamelessly out there and taking notes on the feedback we hear and share with smalltalkers 2. having regular hangouts to expand all that, wildly

Here is another thing we can do:

1. stay like it is now and 2. loose current talent

If we take the first road we don’t know what could happen

If we take the second road, we know what will happen

The problem is not being small, the problem is thinking small and preserving that attitude

So the question is are we willing to review the things that we’re doing that aren’t bringing the results we expect? and what’s the action decided for next week? and the next?

On Apr 29, 2014, at 10:49 AM, Norbert Hartl <norbert@hartl.name> wrote:

Who is people? Is it inside Inria? If not it might be a good idea to make it visible to the public so a few more people could chime in.

Norbert

That question brings us to: Get the Heck Out of the Building in Founder’s School: Part 2 http://steveblank.com/2014/03/18/get-the-heck-out-of-the-building-in-founder... and this: http://www.gv.com/lib/getting-out-of-the-building-isnt-enough

People introduced to amber is highly unlikely to be aware of Pharo but that does not change the fact that if you as Amber tell this coder "hey I also offer a very powerful IDE" instead of "hey I also offer a very basic IDE" it will be obviously a big plus for him. I just dont think diverting resources to creating an IDE for Amber is optimal. Anymore than Pharo ignoring Amber and creating its own complex solution to access Javascript libraries would be optimal. I think that Amber and Pharo should be merged, maybe not as code bases or not as projects or websites or whatever but as communities. The way I see it both work towards the same goal, making the Smalltalk voice heard loud and clear. I think a closer integration is for the benefit of both communities. The communities are too small to afford to spread thin. Inventing the future is all cool and great but this world runs on money , not on progress and innovation. In most cases money is a big obstacle to progress. "I love progress , I just dont like change" (sorry dont remember where I got the quote from) So I think it would be wiser to be practical and take advantage of what you already have and take it one step further at a time. On Tue, Apr 29, 2014 at 2:56 PM, Sebastian Sastre < sebastian@flowingconcept.com> wrote:

On Apr 29, 2014, at 5:58 AM, Markus Fritsche <mfritsche@reauktion.de> wrote:

1. The most kick ass thing I can think of would go something like this...

The most "kick ass thing" I can think of is...

- Have a Pharo World being interfaced to an amber frontend, so you can use your image through your browser.

... on second thought... that might be cool but after five minutes it'll make me think "and now?"

Yes, what now?

I agree the other way around is way more attractive. Both ways is awesome.

But even that is invisible for most people.

What they need is to look like ass kickers in the face of their clients: *people that buys the experience of the app not how easily hard they debug*.

In the other hand, if they can debug better and use this tools to get to the goal faster, that counts as ass kicking sure.

What is happening now is that they’re using the javascript universe of packages and make the right salad.

There is a market opportunity in designers that shyly try to get something non-static done for their clients.

We should be listening to these guys.

Using wordpress and adding plugins is profitable but it has limits and they know that.

So they are using the first thing that’s elegant and work with a conf or two for a start.

Express js is a good fit for them because is easy for them to grasp the basics

What I see is the opportunity of thinking beyond engineers and computer scientists and being an option for designers.

What I don’t see is us being prepared or caring about that huuuuuuuuuuuge, even gigantic market.

If we keep being reactive and defensive on this we will distance ourselves of:

Design Principles Behind Smalltalk The purpose of the Smalltalk project is to provide computer support for the creative spirit in *everyone*.

Let’s say eveyone is too much. Lets strategically focus in the two audiences that are influential the most: 1. computer scientists 2. designers

They are the innovators and disruptors.

They are inventing the future

In which ways are we making their lives easier? (or staying at the margin from that)

I was talking mainly about the IDE and other areas where Pharo and Amber overlap. I think Amber is awesome for what it is and should be kept developed independently. "If your ass is being kicked by AngularJS <https://angularjs.org/> or Meteor<https://www.meteor.com/> because you reach the goal faster using those, your next contract is not to be powered by Amber." is this not the cases if you cannot beat them join them ? Do you seriously believe Amber or Pharo can compete AngularJS , correct me if I am wrong, when behind this framework is Google itself ? "In regard to Pharo, becoming Amber friendly is “attracting javascript people to the smalltalk world” are you joking? is the best marketing strategy Pharo can dream about!" Did I said anything against that ? I agree that is great. "With some add ons Amber can do the things Meteor does in the front end." Thats all great but again I ask why not leverage the power of Meteor.js for Amber and Pharo in the first place and sprinkle it with some smalltalk libraries to bring it closer to Smalltalk philosophy ? Isnt this the whole point of doing OOP, to reuse code ? Why should it be Amber / Pharo vs the Others ? Instead as "Amber a way to make your favorite javascript libraries even easier and more accessible / dynamic" Do you seriously believe people will give JS for Amber alternatives ? No thats not a best marketing strategy.

On Apr 29, 2014, at 10:03 AM, kilon alios <kilon.alios@gmail.com> wrote:

is this not the cases if you cannot beat them join them ? Do you seriously believe Amber or Pharo can compete AngularJS , correct me if I am wrong, when behind this framework is Google itself ?

Against is a no-chance thing. It’s either you do it better or you join on top. I’m curious about where are we going