Project Ramping Up

[ttglennhall]

We are starting to ramp up the San Diego Sweepaware project! Everyone that is watching this project but I have not yet had the pleasure to meet my name is Tiffany Glenn-Hall and I am going to be the tech coordinator. There have been several productive openSD meetings in the past of this project and I can't wait to get to know everyone!

For simplicity and speed of development we talked about making a responsive website app (meaning it will resize to any screen) instead of an iOS/Android app. The website will be hosted on Amazon's AWS through the openSD account. For now that is the direction we are going to go, if you feel strongly that this is the wrong choice, please voice your opinion.

Just last night @slester created our stack so we can start developing. We have decided to go with Node.js and MongoDB. Please respond with any comments, questions, or concerns for these choices.

In terms of front end development I was thinking we could use Boostrap, there is a similar website for Denver that we can look at for our initial work. It does not have maps, and the visuals created by @jerryHall and the May meetup group were much more visually appealing.

This was more of an informational issue to get the conversation started but I am also going to start other issues that people can sign up for and let us know what they are working on. I am so excited to get the project moving!! Thank you to everyone who has contributed in the past and are coming back to complete this excellent idea!

Please respond with comments, questions, or concerns for this project. Also please feel free to introduce yourself, what you have experience with, and perhaps where you would like to gain experience.

[ttglennhall]

In order for us to have conversations, without improperling using the issues board we have added a commentary website.

https://opensd.slack.com/

Please open this website and sign up so we can more easily communicate.

[slester]

@ttglennhall You actually need to be invited by the creator in order to have access.

[ttglennhall]

Oh, then everyone who wants to be involved in all discussion please e-mail me (ttglennhall@gmail.com) your e-mail address.

[cgiovando]

Hey @ttglennhall and @slester - thanks so much for taking the lead on this project. Maybe it was not mentioned by Jerry last night, but we actually already use a Slack-like chat platform called Gitter, which is more open and does not require invites. Links to each project's chat room are at the top of readmes / homepages

The one for sweepaware is https://gitter.im/opensandiego/sd-sweepaware

And the general Open San Diego room is https://gitter.im/opensandiego/general

Everyone can read public gitter rooms and you only need a Github account to join the conversation. It's free and there are no search limits for all public open source projects. It also has the same type of integrations like Slack.

BTW, we already have an Open San Diego slack at https://opensandiego.slack.com but we kinda abandoned that once we discovered Gitter ;-)

[ttglennhall]

Thanks @cgiovando it sounds like Gitter is a better bet.

[hvanbakel]

Hi,

I just wanted to let you know I've created a service that does exactly what this project is aiming for. Free of charge as well. It's over here: http://sweepingalerts.com/ca/sandiego

I'm hoping to make a more formal announcement soon.

[jerryhall]

This is sweet Hans.

I've been out of touch the last months and would love to catch up.

Do you want some help promoting this site to the greater San Diego?

Jerry

Jerry Hall CivicArchive

Jerry@CivicArchive.com 858-344-1104 cell

LinkedIn http://Linkedin.com/in/enjoypb

On Sat, Apr 16, 2016 at 10:19 AM, Hans van Bakel notifications@github.com wrote:

Hi,

I just wanted to let you know I've created a service that does exactly what this project is aiming for. Free of charge as well. It's over here: http://sweepingalerts.com/ca/sandiego

I'm hoping to make a more formal announcement soon.

— You are receiving this because you were mentioned. Reply to this email directly or view it on GitHub https://github.com/opensandiego/sd-sweepaware/issues/7#issuecomment-210861110

[hvanbakel]

I've set the launch for sweepingalerts to next Tuesday the 19th. Any help in promoting the site would be greatly appreciated!

[jerryhall]

Cool! On Apr 16, 2016 8:22 PM, "Hans van Bakel" notifications@github.com wrote:

I've set the launch for sweepingalerts to next Tuesday the 19th. Any help in promoting the site would be greatly appreciated!

— You are receiving this because you were mentioned. Reply to this email directly or view it on GitHub https://github.com/opensandiego/sd-sweepaware/issues/7#issuecomment-210946041

[cgiovando]

Sure @hvanbakel - can you post a link to the repo?

[hvanbakel]

It's not (yet) on a public repo. I've got a private account at Github, at this time I still need to do some cleaning up on the code. All the code runs on Azure, I need to find a way to keep the account details out of the code as those link to my account.

On Sunday, April 17, 2016, Cristiano Giovando notifications@github.com wrote:

Sure @hvanbakel https://github.com/hvanbakel - can you post a link to the repo?

— You are receiving this because you were mentioned. Reply to this email directly or view it on GitHub https://github.com/opensandiego/sd-sweepaware/issues/7#issuecomment-211108597

[bcipolli]

I need to find a way to keep the account details out of the code as those link to my account.

@hvanbakel The typical way I've seen this done is:

1. Access the account details from a config file
2. Create an example config file (e.g. .config-example) with a different filename than the actual config file, and will all variables... with dummy values.
3. In your README.md setup instructions, tell the user to copy that example file to the real location, and to replace the appropriate variable values.

[hvanbakel]

I understand, I'm also waiting for another feature within Azure however. Internally it uses message queues to created disconnected subsystems. However, at the time of writing I cannot schedule azure to add a message to a specific queue at a given time, therefor this is handled by a crafted HTTP request. This should be fixed in the near term.

On Sunday, April 17, 2016, Ben Cipollini notifications@github.com wrote:

I need to find a way to keep the account details out of the code as those link to my account.

@hvanbakel https://github.com/hvanbakel The typical way I've seen this done is:

1. Access the account details from a config file
2. Create an example config file (e.g. .config-example) with a different filename than the actual config file, and will all variables... with dummy values.
3. In your README.md setup instructions, tell the user to copy that example file to the real location, and to replace the appropriate variable values.

— You are receiving this because you were mentioned. Reply to this email directly or view it on GitHub https://github.com/opensandiego/sd-sweepaware/issues/7#issuecomment-211109000

[bcipolli]

Is this a reason not to open-source the project? I don't get the reasoning.

@jerryhall does CfA help support projects that are not open-source?

On Sun, Apr 17, 2016 at 1:40 PM, Hans van Bakel notifications@github.com wrote:

I understand, I'm also waiting for another feature within Azure however. Internally it uses message queues to created disconnected subsystems. However, at the time of writing I cannot schedule azure to add a message to a specific queue at a given time, therefor this is handled by a crafted HTTP request. This should be fixed in the near term.

On Sunday, April 17, 2016, Ben Cipollini notifications@github.com wrote:

I need to find a way to keep the account details out of the code as those link to my account.

@hvanbakel https://github.com/hvanbakel The typical way I've seen this done is:

1. Access the account details from a config file
2. Create an example config file (e.g. .config-example) with a different filename than the actual config file, and will all variables... with dummy values.
3. In your README.md setup instructions, tell the user to copy that example file to the real location, and to replace the appropriate variable values.

— You are receiving this because you were mentioned. Reply to this email directly or view it on GitHub < https://github.com/opensandiego/sd-sweepaware/issues/7#issuecomment-211109000

— You are receiving this because you commented. Reply to this email directly or view it on GitHub https://github.com/opensandiego/sd-sweepaware/issues/7#issuecomment-211109637

[hvanbakel]

At this point it's a purely technical problem, this is a current limitation of Azure which forces me to use a way that is not secure. This is for scheduling purposes only, no data is being exposed. Open sourcing at this time would make the system vulnerable. This is not how it's supposed to be, but that's something that azure needs to fix.

The basic problem is simple, I use a message queue to communicate with a disconnected system. A scheduler should push a message directly to the queue, however, that's not possible so it calls a public endpoint which is just security by obscurity.

At this time there's no need for any assistance code wise, if someone wants to help out to have people get to know the service, that's greatly appreciated. I'm planning on testing the waters first. If this is a real problem and a lot of people sign up it makes sense to keep building on the system. For that purpose, not being able to open source now doesn't make it less useful (or not). At this time all of the accounts are my personal accounts, costs associated with that are on me as well, this was just a choice to get things up and running (and I didn't know this project even existed up until last week).

On Sunday, April 17, 2016, Ben Cipollini notifications@github.com wrote:

Is this a reason not to open-source the project? I don't get the reasoning.

@jerryhall does CfA help support projects that are not open-source?

On Sun, Apr 17, 2016 at 1:40 PM, Hans van Bakel <notifications@github.com javascript:_e(%7B%7D,'cvml','notifications@github.com');> wrote:

I understand, I'm also waiting for another feature within Azure however. Internally it uses message queues to created disconnected subsystems. However, at the time of writing I cannot schedule azure to add a message to a specific queue at a given time, therefor this is handled by a crafted HTTP request. This should be fixed in the near term.

On Sunday, April 17, 2016, Ben Cipollini <notifications@github.com javascript:_e(%7B%7D,'cvml','notifications@github.com');> wrote:

I need to find a way to keep the account details out of the code as those link to my account.

@hvanbakel https://github.com/hvanbakel The typical way I've seen this done is:

1. Access the account details from a config file
2. Create an example config file (e.g. .config-example) with a different filename than the actual config file, and will all variables... with dummy values.
3. In your README.md setup instructions, tell the user to copy that example file to the real location, and to replace the appropriate variable values.

— You are receiving this because you were mentioned. Reply to this email directly or view it on GitHub <

https://github.com/opensandiego/sd-sweepaware/issues/7#issuecomment-211109000

— You are receiving this because you commented. Reply to this email directly or view it on GitHub < https://github.com/opensandiego/sd-sweepaware/issues/7#issuecomment-211109637

— You are receiving this because you were mentioned. Reply to this email directly or view it on GitHub https://github.com/opensandiego/sd-sweepaware/issues/7#issuecomment-211109946

[jerryhall]

I know Hans has been developing this on his own and it's up to him to OS it or not. I can't imagine CfA not supporting non-OS stuff as they do all the time through their incubated companies and others they support. I'm just glad it's being done and that Hans wanted to run with it since our Brigade hasn't taken the steps yet.

Jerry

Jerry Hall CivicArchive

Jerry@CivicArchive.com 858-344-1104 cell

LinkedIn http://Linkedin.com/in/enjoypb

On Sun, Apr 17, 2016 at 1:54 PM, Hans van Bakel notifications@github.com wrote:

At this point it's a purely technical problem, this is a current limitation of Azure which forces me to use a way that is not secure. This is for scheduling purposes only, no data is being exposed. Open sourcing at this time would make the system vulnerable. This is not how it's supposed to be, but that's something that azure needs to fix.

The basic problem is simple, I use a message queue to communicate with a disconnected system. A scheduler should push a message directly to the queue, however, that's not possible so it calls a public endpoint which is just security by obscurity.

At this time there's no need for any assistance code wise, if someone wants to help out to have people get to know the service, that's greatly appreciated. I'm planning on testing the waters first. If this is a real problem and a lot of people sign up it makes sense to keep building on the system. For that purpose, not being able to open source now doesn't make it less useful (or not). At this time all of the accounts are my personal accounts, costs associated with that are on me as well, this was just a choice to get things up and running (and I didn't know this project even existed up until last week).

On Sunday, April 17, 2016, Ben Cipollini notifications@github.com wrote:

Is this a reason not to open-source the project? I don't get the reasoning.

@jerryhall does CfA help support projects that are not open-source?

On Sun, Apr 17, 2016 at 1:40 PM, Hans van Bakel < notifications@github.com javascript:_e(%7B%7D,'cvml','notifications@github.com');> wrote:

I understand, I'm also waiting for another feature within Azure however. Internally it uses message queues to created disconnected subsystems. However, at the time of writing I cannot schedule azure to add a message to a specific queue at a given time, therefor this is handled by a crafted HTTP request. This should be fixed in the near term.

On Sunday, April 17, 2016, Ben Cipollini <notifications@github.com javascript:_e(%7B%7D,'cvml','notifications@github.com');> wrote:

I need to find a way to keep the account details out of the code as those link to my account.

@hvanbakel https://github.com/hvanbakel The typical way I've seen this done is:

1. Access the account details from a config file
2. Create an example config file (e.g. .config-example) with a different filename than the actual config file, and will all variables... with dummy values.
3. In your README.md setup instructions, tell the user to copy that example file to the real location, and to replace the appropriate variable values.

— You are receiving this because you were mentioned. Reply to this email directly or view it on GitHub <

https://github.com/opensandiego/sd-sweepaware/issues/7#issuecomment-211109000

— You are receiving this because you commented. Reply to this email directly or view it on GitHub <

https://github.com/opensandiego/sd-sweepaware/issues/7#issuecomment-211109637

— You are receiving this because you were mentioned. Reply to this email directly or view it on GitHub < https://github.com/opensandiego/sd-sweepaware/issues/7#issuecomment-211109946

— You are receiving this because you were mentioned. Reply to this email directly or view it on GitHub https://github.com/opensandiego/sd-sweepaware/issues/7#issuecomment-211115038