eweekchat

A3: is that a technology question?

It's about IT security, so yes

A3. I believe IT (when not faced with legal obstacles/hoops) can definitely help but this is just not a technical problem. It is a culture and influence issue. Security isn't just about what happens at the ballot box but everything in the months before it.

A3: Completely means 100%... is any security 100%?

Not as long as the decision makers ask questions like this: https://www.nbcnews.com/video/senator-asks-how-fac...

@jmeserve Great point!

A3 : From a technology perspective sure we can. Will anyone ever trust that we can... likely not.

A3: ​​​​​​​“Ever” is a long time. But if we do, it won’t be by #digitally replicating the ballot box process. Technologies like #blockchain might help – but real success requires a dramatic rethink. #eWEEKchat

A3: This election cycle will prove pivotal in shaping the future of election security, that's for sure!

A major challenge to solving this is the trade-off between security and usability which is a law of nature.

@gssor Trust isn't always a technical thing either but a warm-and-fuzzy feeling for most voters out there. I think we can but certain large firms (e.g., FB) have cast the entire industry in a negative light and that makes it very hard to establish trust with citizens.

A3: no, there is no such thing as complete security. Even disconnected and unplugged, still hackable. What we need are low-tech hand-marked paper ballots with audits, etc but even then we need to secure and validate/audit the count transmission

A3: One of the key IT terms we seldom hear applied to the election process is "redundancy." Yet, a recent flaw did not cause a problem because it featured a redundant print to paper ballots that could be counted and certified when the data was demonstrated to be flawed.

@christo4ferris Yes, security (and defense) in depth is a great idea. #Blockchain isn't going to solve it (Yes, I said it).

Unclear to me that the real problem we have had in elections is "security" in the traditional sense (did only the appropriate people vote, was it counted correctly, etc). Most of the talk has been about outside parties "influencing" the election

@ASGTec as someone deeply involved in #blockchain, I can say that it might help, but should never be seen as panacea. Could be used to create an audit log that can be validated.

@nirajtolia Right, trust extends way beyond this use case. Frankly we still see incredibly backwards views of privacy and security around cloud infrastructure that are just wrong.

@nirajtolia completely agree.

Hey, Welcome, Simon!

A2: true (i.e. important and not a passing fad).

HUGE TREND. Yes, I am biased because I work on it but I also try to separate hype from reality. Still early days but all the customer deployments point in the same right direction.

A2: yuge, even... projects such as Kubeflow likely to gain significant traction

There are folks service TBs of day/day out of #Kubernetes on apps that you probably love and use. Given the velocity of app dev in this environment and how much easier stateful applications have become, this isn't a fad. Next Linux or vSphere likely.

@christo4ferris That is a great point! #Kubernetes seems to have won the DL market and we will see even more data-intensive frameworks get replatformed on this in 2020.

A2. #Data orchestration is part of rebuilding the #infrastructure and #dataarchitecture. It's important, but it's just one part of rebuilding the data #architecture. #eweekchat

Isn't it just a part of the plumbing, though--albeit very important?

well, yes, but all IT infrastructure is just part of the plumbing, depending on your perspective.

Yes - and the plumbing is being reinstalled! #eWEEKchat

A1. False. Apart from false advertising from some providers (looking at you 5GE), it is still early days for both 5G devices and apps that can leverage it. That said, 5G is quite exciting to me!

Based on what I'm hearing it will take somewhat longer to achieve sufficient coverage and sufficient cost-settling, so perhaps the next two years.

A1 : True - assuming the device manufacturers embrace the new tech quickly in 2020

A1: false. It will take a while for 5G to roll out, and the devices do so much more than what the network enables or doesn't. So my guess is we're looking at more like a 3 year rather than 1 year horizon.

@nirajtolia agree - don't think the national rollout will be enough to compel the mainstream to buy 5G phones just for the sake of 5G in 2020

T-Mobile is making a big deal of rolling out its national network NOW.

@gssor Given hardware timelines (ask me about shipping a "box" over drinks!), if it isn't in their pipeline already, vendors won't be shipping them this year at scale.

T-Mobile is going to have it's hands full merging with Sprint. Letting their marketing get ahead of their tech.

Marketing/Hype vs. Reality.

We are going to hear so much about 5G though that we'll be sick of it in 2020. Now that's a true prediction.

@nirajtolia Exactly

A1. False! There is still a lack of coverage and economic considerations to take into account. Early adopters will jump in but the rest of us will wait for coverage. #eweekchat

A major issue with millimeter waves is the signal propagation, and only time will tell whether providers are prepared to invest heavily enough in densification to deliver on the promises.

That said, folks in my neighborhood are all up in arms about 5G rollout on a pole next to them. A lot of NIMBY going to here in Northern CA.

@nirajtolia and probably the same folks that complain about crappy cell service.

@nirajtolia Fair enough point... But other parts of the world are well on their way, with deployments for over a year already. And phones are only one perspective.

@gssor Agreed - IoT signal persistence is a huge application-in-waiting. Apps that require zero latency.