Now that Amazon Web Services has shut down Parler, will the social network rebuild elsewhere? Parler CEO John Matze says Parler will be back with “many competing for our business.”
Don’t bet on it. There appears to be no major public cloud or hosting company willing to give Parler a home. But, even if there were, Parler would find it almost impossible to return anytime soon. Parler will need to build its own infrastructure.
Indeed, Matze afterward admitted on Fox News that no one wants to work with Parler after Amazon dropped the company. Bringing Parler back up is “basically impossible,” Matze concluded.
Even if they did, as Corey Quinn, Chief Cloud Economist at the Duckbill Group and who covers AWS like paint, explained in a series of tweets that it can’t be done:
Getting booted off of AWS is virtually unheard of; when people leave intentionally the planning takes months; execution can take years. It’s a lot harder than you think for a few reasons.
It takes time to move all of that data (there’s a reason downloads take a while), but that’s just the beginning. AWS doesn’t just sell “big empty computers.” They offer higher-level services. Preconfigured databases, automatic video streaming, etc.
Parler claims they didn’t use these higher-level services. Taking that as true, there are still problems. The way AWS’s services work–how you create them, how long that creation process takes, how you get data onto them? They behave differently in AWS’s world than elsewhere.
A lot of assumptions about how the servers behave are “baked in” to how Parler (and any AWS hosted application) are built. A lot of companies don’t realize that those assumptions are there until they try to move.
That’s why migrations take months/years. Parler has 30 hours.
Had 30 hours. The site is now down.
Setting up a social network itself is easy. Many web-hosting companies offer easy-to-setup open-source networks such as Elgg, diaspora*, and Okuna. With these, you can set up a small social network in an afternoon. Setting up a social network that can handle more than 8 million Parler users is another matter entirely.
It’s not just a matter of money. Multi-millionaire conservative Parler backer Rebekah Mercer, daughter of hedge-fund investor and Cambridge Analytica co-founder Robert Mercer, has ample funds. By the time it closed, Parler was costing about $300,000 a month to AWS alone to run.
The question is: Will Mercer even want to continue to invest in a business that has no real business plan? CEO Matze has admitted, “Every vendor from text message services to email providers to our lawyers all ditched us too on the same day.”
Parler doesn’t have the in-house expertise to make the migration. Parler founder and lead engineer Jared Thomson has two years of full-time tech experience outside of the social network and none of that is with cloud computing. Alexander Blair, Parler CTO, is a self-described Linux system administrator with five years of system deployment and management work experience. In total, Parler has 35 employees, who were already stressed dealing with the network’s growth after the election. This job would push the limits of far larger and more cloud-savvy staff than Parler can boast.
Some companies, such as VanwaTech — aka OrcaTech, a Vancouver, Washington hosting company — might host a new Parler. Previously, VanwaTech provided a home for the neo-Nazi The Daily Stormer; 8kun, previously 8chan; and numerous QAnon conspiracy sites. VanwaTech, however, also does not have the technical expertise to migrate Parler.
Simply moving data is only the start of the answer. The Parler server program must be transferred to other cloud-based servers. Parler runs on proprietary closed-source software running on Linux. As a closed-source program, it will be far harder to port to another platform.
Parler’s data was kept on Amazon S3 servers. S3 is a popular and well-documented object storage service. Simply moving data from it is easy. Using that data on an updated proprietary program on a new cloud is another matter entirely.
To deliver its content to users will also prove troublesome. For its front-end, Parler used Amazon Cloudfront. This is AWS’s content delivery network (CDN). Without a CDN, a revived Parler would have a difficult time delivering its content to users in a timely fashion.
A new Parler would also have to deal with setting up Distributed Denial of Service (DDoS) protection. Major DDoS companies such as Cloudflare have no interest in protecting Parler. Still, DDoS protection for controversial sites is essential. Cloudflare dumping 8chan was a major reason why the site almost went under. Today, VanwaTech-based sites rely on DDoS-Guard, a Russian DDoS protection company.
Put it all together and I doubt very much we’ll see a revival of the old Parler. I expect that a new right-wing social network will emerge.
This is not just a problem for Parler. Any company that wants to build a site or social network and relies on AWS or any of the other major public clouds potentially faces the same problems. Generally speaking, AWS wants your business, but at the end of the day, as Parler discovered, AWS’s terms of service make it possible for them to shut down any site.
If you really want to make sure your site or service stays up, you must still run them on your own hardware in your own datacenter. Other controversial sites, such as Pornhub, have successfully used this model. To be certain your website or social network stays up no matter what, the private cloud model is still the best one.