Forgot your password?
typodupeerror

Comment: Re:backup for 911 (Score 1) 115

by aaarrrgggh (#48203849) Attached to: Software Glitch Caused 911 Outage For 11 Million People

Which is kind of ok as long as your problem happens in that area, and there is no actual natural or man-made disaster in play.

A 911 dispatch center has much better resources than a local police or fire department main desk. If you are lucky, they can operate at 10% of the capacity as a proper dispatch center.

The system should be more robust. It has improved dramatically since 9/11 with absurd amounts of cash poured into many facilities, but it can't do everything. An alternative solution is to not expect 9/11 to be robust, but hard to tell if that is a good approach for most communities.

Comment: Re:Boston (Score 1) 173

Deploying in a market is more likely a 20-year payback, and incremental customers are a 2-year payback. The benefit to Verizon is in deprecating copper infrastructure, which has high maintenance costs due to age, but overlaying new fiber pathways on the existing copper pathways is expensive.

They want to hit markets where they can have an incremental income from FIOS, and where the chances of competition are low. That means prioritizing affluent communities that have the wherewithal to create alternatives if the need isn't addressed.

Comment: Re:If only (Score 1) 99

by aaarrrgggh (#48201545) Attached to: Internet Broadband Through High-altitude Drones

Towers fail in time to deploy. Even small scale towers- utility poles or street lights) in mesh networks take substantial time to get licensing and rights-of-way in place. Wireline services have similar constraints, but just for the initial pathway installation, which can be leveraged for decades. A wireless technology upgrade needs to happen every 2-5 years if the service is successful.

Comment: Re:I'm betting on balloons (Score 1) 99

by aaarrrgggh (#48201527) Attached to: Internet Broadband Through High-altitude Drones

There was an article a few years back on /. about a network of weather balloons with hotspots, forget the name of the system. It is a bit labor intensive, since each unit only stays in flight 3-5 days, and the shoebox needs to be recovered and returned.

Zeppelin ROVs have also been discussed on /. before, and offer substantial improvements over the system, but not really lower costs-- just trade offs.

Traditionally, wireless always makes sense when you have limited subscriber density or portability requirements. Where it usually fails is that it can't scale as quickly when bandwidth needs increase. As an old codger, I will also recall the discussions of MetroLink's Ricochet Wireless system here on /. 15-17 years ago-- same basic concepts, same challenges.

Wireless is always great for yesteday's use cases. Time to market from yesterday is what kills them, which airborne solutions can improve on if done right... but I won't hold my breath for low latency gigabit connections from a UAV.

Comment: Re:"Fear" (Score 1) 383

If you don't want to believe what someone is saying is the truth, you will not trust them. Your opinion is set from sources you do trust, which does not necessarily have your best interests at heart. Politics- next election; news- next ad; military- next budget; church- blind following; nobody is impartial.

What you can control is your own behaviors to ensure the people you trust have common goals that are important to you. A Talking Head can never be in that position. Seek out experts...

Comment: Re:Boston (Score 4, Insightful) 173

Verizon isn't seeing the return on capital for FIOS; that is well known. They think they can increase subscriber rates in areas they have covered and recover the capital that way.

What they completely miss is the fact that the use-cases that will drive more valuable service plans only exist when ~gigabit networks are available everywhere.

The problem I see with either approach is that business internet costs aren't going down fast enough to push that evolution. You get better speeds for less in a co-lo, but that doesn't help enough if you use a single office location.

Comment: Depends on if you want to get calls... (Score 4, Informative) 105

by aaarrrgggh (#48184741) Attached to: Ask Slashdot: LTE Hotspot As Sole Cellular Connection?

If you don't need to receive calls, you can control use pretty well and maybe make it work for slightly less money.

Speaking from personal experience while traveling though, it is really a pain, especially for a prolonged period of time. If you have a specific need to use multiple devices for non overlapping functions (laptop, phone, tablet) where the functions really can't be done on a single device then the MiFi is cheaper than getting three SIMs. The only time I broke down and went this route in the last 10 years was in Sydney, where the hotel charged around $25 for wifi, and I only had one Australian SIM card that would work.

Convenience or cost...

Comment: Re:You don't need the bandwidth (Score 1) 150

by aaarrrgggh (#48148703) Attached to: If Your Cloud Vendor Goes Out of Business, Are You Ready?

Lease payments, power, bandwidth are all expensive in a colo. If you are out of cash, you are in trouble. Your assets could be used to wind down operations gracefully... Or to your point to repay your creditors.

Hopefully Google, Amazon, or Microsoft would give reasonable notice of a pending cancellation of a service, but there are no guarantees.

Comment: Re:Local Backups (Score 1) 150

by aaarrrgggh (#48148669) Attached to: If Your Cloud Vendor Goes Out of Business, Are You Ready?

Removable drives work well to a point, but you need at least 3 for a proper rotation. You also have the issues of failure rate and potential for theft, which eventually drive you to the SAN/NAS route which gets expensive quickly. Cloud services for our business have a sweet spot around 4TB of live data; businesses with highly distributed workforces may nap have a much lower threshold.

Cloud services seem to make sense to me for small companies not wanting to invest in servers and to minimize consulting, or companies of any size that just don't want to bother with it, and are willing to pay more for fewer distractions. Out of sight (site) out of mind. Buyer beware...

Comment: Re:Legally binding? (Score 2) 150

by aaarrrgggh (#48148579) Attached to: If Your Cloud Vendor Goes Out of Business, Are You Ready?

Moreover, if a big cloud vendor quickly closes shop, interdependencies and network effects are likely to have an impact on your contingent vendor.

We have hosted email, and will likely move to Amazon Glacier for DR backups; we have local snapshot backups that give all the information locally that would go to Glacier; it is just the earthquake/sprinkler/sabotage scenario that offsite would protect us against, and Glacier is starting to get competitive for our needs.

Like everything, it is a scenario you need a plan for. Depending on the impact, the plan needs to be developed, tested, and re-validated as appropriate.

Comment: Re:Developer unhappiness or Marketshare loss? (Score 1) 229

by aaarrrgggh (#48146597) Attached to: The Subtle Developer Exodus From the Mac App Store

AutoCAD LT is $800.
I think Omni Group had something on there for $140 a while back.
Generally speaking I don't have a need for many $5 apps on my Mac; if that is all it is worth I can use a spreadsheet or something for the functionality. The impulse buys just aren't there.

This is a good time to punt work.

Working...