Small Grant: CloudBridge – UPDATED

[DappMentors]

Introduction

Project Name: CloudBridge

Organization: Dapp Mentors

---

Background

Dapp Mentors is a two-time Sia Foundation grant recipient specializing in developer tooling and decentralized infrastructure. Our previous grants delivered SiaLearn and SiaPeopleLearn, both built on Renterd.

This proposal is a deliberate, focused step. Following the Committee’s direction and the Foundation’s current funding focus, CloudBridge will be built directly on the indexd + Sia Storage SDK pathway, while delivering something useful for everyday users right now.

---

Describe Your Project

CloudBridge is a self-hostable, open-source web application that gives users a clean graphical interface to migrate and back up data from Google Drive and local filesystem directly into Sia using the Sia Storage SDK and indexd.

CloudBridge wraps the indexd + SDK stack in a simple browser-based UI. Users connect to their indexer endpoint once, pick files from Google Drive or their local machine, and start secure streaming transfers without needing to install or operate command-line tools.

Today, getting files into Sia mostly still involves technical hurdles that block non-developers. CloudBridge reduces this to a guided flow inside a browser:

• Connect to your indexer once
• Select files visually
• Start transfer with one action

This is scoped as a minimal working proof of concept within a small grant. The goal is to prove the data migration use case is real, show that our team can deliver reliably on Sia’s current tools, and validate demand for a simplified user-facing layer.

Core Capabilities:

• Indexer endpoint setup with connection validation

• Google Drive and local filesystem connectors

• Simple file browser to preview and select what to transfer

• Streaming uploads via the Sia Storage SDK with no disk-side plaintext staging

• Background transfers with basic status tracking

• Simple job persistence using a local JSON file

• Object pininig/unpinning via indexd, visible on dashboard

• One-click download button for uploaded files

• Docker-based one-command self-hosted deployment

---

Evidence of Demand

CloudBridge targets a real and documented gap: there is currently no simple, open-source web UI built natively on the Sia Storage SDK and indexd that lets non-technical users migrate their data into Sia without touching a terminal.

• Sharing and migrating large files: Moving large datasets, media libraries, and backups remains a core need for users evaluating Sia.

• Avoiding vendor lock-in: Platforms like SproutVideo charge per-seat pricing, while tools like Pixeldrain lack open-source extensibility. Users actively look for open-source alternatives they can control.

• Community validation: Self-hosting communities continue to discuss backup, migration, and storage portability challenges, especially for users leaving centralized providers.

• Ecosystem gap: The Sia Storage SDK and indexd exist and work well, but there is no simple, open-source web app built specifically to onboard everyday users into this flow.

---

Target Users

• Business owners and non-technical decision-makers

• Content creators (filmmakers, educators, digital artists)

• Self-hosting enthusiasts who prefer UI-driven tools

• Developers exploring Sia but not yet comfortable with SDK workflows

• Privacy-focused everyday users

---

How Does This Serve the Foundation’s Mission?

The Committee has made it clear that migrating and sharing large files without vendor lock-in is a top priority, and that the indexd + SDK pathway is the current direction for funded projects. CloudBridge lowers the barrier for users who want to move their data into Sia but are currently blocked by tooling complexity.

Each successful transfer represents real user data moving from centralized systems into user-owned, decentralized storage on the Sia network.

---

Compliance Questions

Are you a resident of any jurisdiction on that list? No

Will your payment bank account be located in any jurisdiction on that list? No

---

Grant Specifics

Amount Requested: $8,500 USD
Timeline: 2 months

Category	Cost (USD)	%	Justification
Core backend & SDK/indexd integration	$5,300	62.35%	Sia Storage SDK integration, indexd pinning, streaming transfer engine, job queue
Frontend UI/UX	$2,000	23.53%	File browser, transfer dashboard, settings, and onboarding
Testing, QA & documentation	$600	7.06%	End-to-end tests, error handling, README, and self-hosting guide
Infrastructure & DevOps	$600	7.06%	Docker setup, CI pipeline, GitHub repo configuration
Total	$8,500	100%

---

Goals and Timeline

Month 1: Foundation & Connectors

Due: June 30, 2026 (Repo setup, indexer connection, connectors, and transfer engine)

• GitHub repo is public with CI pipeline, basic README, and open-source license

• Users can add and validate their indexer endpoint connection via the configuration module

• Google Drive connector lets users browse and select files/folders

• Local filesystem connector supports folder and file selection

• Selected files are uploaded to Sia via the SDK streaming pipeline with live status shown in the UI

• Uploaded objects are pinned via indexd and visible in the dashboard

• Basic transfer logs are visible per job

Month 2: Stability, Polish & Release

Due: July 30, 2026 (Logs, persistence, onboarding, testing, and docs)

Weeks 1-2: Remaining Development

• Simple job persistence is operational so transfer status survives restarts

• Basic error handling and retry for failed transfers are functional

• Guided onboarding flow from indexer setup to first transfer is complete and functional

Weeks 3-4: Testing, Documentation & Release

• End-to-end test suite passes for both Google Drive and local filesystem flows

• Failure scenarios tested: invalid indexer endpoint, connection drops, interrupted transfers

• Docker Compose deployment tested on a clean Ubuntu server

• Complete self-hosting guide published in the repo

• Technical documentation covering setup, configuration, and usage

• v1.0 release tagged on GitHub with clear usage instructions

---

Post-Grant Plan

CloudBridge will be our first delivered tool built natively on the Sia Storage SDK and indexd. A successful v1.0 means real users can move files from Google Drive into Sia without touching a terminal, and we will have the deployment data and feedback to back it up.

Future work will be evaluated based on real usage, feedback, and demonstrated demand. Any follow-up proposal will be grounded in feedback from actual users and deployments.

---

Risks

• Sia Storage SDK or indexd API changes: Mitigated through an abstraction layer that isolates SDK calls from core application logic, allowing targeted updates without major rewrites.

• Google Drive API changes: The Google Drive API is mature, so major breaking changes are unlikely within a 2-month window.

• Indexer availability: Mitigated by configurable fallback and retry mechanisms within the upload queue.

---

Development Information

Open-source: Yes. All code will be publicly licensed from day one.

Repository: CloudBridge Repo

Agreement: Yes, we agree to submit monthly progress reports using the official Sia Foundation template.

Contact:

Email: [email protected]

LinkedIn: https://www.linkedin.com/in/darlington-gospel

[pcfreak30]

Question. It has been made clear that the ecosystem is moving towards indexd, and that renterd based applications will not be accepted.

What of these facts have you confused? This request, as written.. is DOA.

The committee response was referring to s3d, as far as I can tell, and that actually isn’t ready yet anyways so you would want to wait until a release on that repo is tagged.

Thanks.

[DappMentors]

Hey @pcfreak30, appreciate the input.

CloudBridge is proposed directly on the pathway the Committee recommended in the SiaTrans rejection feedback. If that direction has changed, I’m confident the Committee will let me know.

[pcfreak30]

They never said anything about renterd in that response, you made that assumption. They said s3 and rclone…

[DappMentors]

Hey @mike76, quick question on CloudBridge.

The Committee’s feedback recommended “S3 and Rclone” but didn’t specify which S3 implementation. Would the Committee prefer Renterd S3, s3d, or Indexd + SDK for something like this?

Just looking for a clear steer so I can make sure we’re building on the right path. Happy to adjust the proposal accordingly.

Thanks.

[CtrlAltDefeat]

Have you personally used Sia for backups yet? If so, what concrete issues did you run into?

Right now, the “Evidence of Demand” section doesn’t clearly show actual evidence. It reads more inferred than demonstrated, and on top of it more aimed at Sia in general than at the proposed tool.

I’m also not entirely convinced that existing tools like rclone are a significant enough barrier to require an additional layer. If this functionality is needed, would it make more sense to build on top of rclone or contribute upstream rather than creating a separate tool?

Rclone has a built-in GUI and simple workflows. For example, a one-liner can sync a folder, or you can mount the source and destination as local directories for drag-and-drop use.

Regarding the target users: In many cases, those users wouldn’t be the ones setting this up directly, or could already rely on existing abstractions as explained above.

Regarding the mission; You’re implying a need to lower existing barriers. Do you have any proof these barriers even exist, or are as high as you claim?

Assuming the described problem exists; how are you sure the proposed is what will remove the friction?

[mecsbecs]

Hi @DappMentors - jumping in to clarify that building on renterd is no longer a funding focus (more detail can be found in the new funding guidelines) so it should not be included in this proposal.

@CtrlAltDefeat raises some solid points and questions that I would like to see answered.

[DappMentors]

Hey everyone, thanks for the candid feedback.

@mecsbecs, I appreciate the clarity on the new funding guidelines.

To give some context, I proposed the S3/Rclone stack because the committee explicitly recommended starting there in my SiaTrans feedback… And the natural available pathway to that end was to use Renterd.

The ecosystem direction toward indexd + SDK is clear and great, and I’m more than happy to pivot. My original architectural preference was to build directly on indexd + SDK anyway.

@CtrlAltDefeat, you raise fair points about rclone having built-in workflows. It’s useful for power users, but it still requires installing rclone, running commands to start it, configuring remotes, and keeping things running in the background. For non-technical folks (content creators, small business owners, regular users), that remaining friction is exactly what stops them from trying Sia.

CloudBridge aims to be a zero-friction web UI for people who just want a familiar Google Drive style experience without touching a CLI.

I will revise the CloudBridge proposal to strip out the Renterd dependency and align it strictly with indexd + SDK, while keeping the scope small and focused as the committee initially requested.

I will update the thread once the new draft is ready.

[pcfreak30]

@DappMentors

And the natural available pathway to that end was to use Renterd.

Again you made that assumption . And right now s3d isn’t exactly ready so your grant request might have to wait until that is usable.

[CtrlAltDefeat]

@DappMentors

@CtrlAltDefeat, you raise fair points about rclone having built-in workflows. It’s useful for power users, but it still requires installing rclone, running commands to start it, configuring remotes, and keeping things running in the background. For non-technical folks (content creators, small business owners, regular users), that remaining friction is exactly what stops them from trying Sia.

As per previous feedback; this is still a claim without backing.
Please respond to all feedback I provided.

[DappMentors]

Hey @CtrlAltDefeat,

Fair point, let me be more concrete.

On personal experience: Yes, I’ve used Sia extensively for backups and real applications. Through our previous grants I created the Sia Blockchain tutorial playlist, where I built and taught a full Web3 movie streaming dApp (VidTV) using Sia for media storage and streaming.

I guided learners through the entire process: setting up storage, uploading files, integrating it into an app, and hitting the practical pain points. That’s where the biggest friction became obvious. Even motivated developers and creators often get stuck at the “how do I actually get my files in here without fighting the terminal?” stage.

On evidence of demand: The usability gap shows up consistently in those training videos and in self-hosting communities. People want to move their Google Drive libraries or media folders to a private and self-hosted alternative, to capture this on the Sia network, using the (Renterd/S3 + Rclone; even with its web GUI) path still requires too many setup steps. CloudBridge is meant to be the simple browser layer on top: connect once, pick files, transfer. Not a replacement for Rclone, but a friendlier onboarding tool.

On target users & whether this actually removes friction: I hear you that some users won’t set things up themselves. But many content creators and small business owners do manage their own archives. For them, going from “Google Drive style” to a clean web UI that handles the whole migration is a much lower bar than learning remotes, mounting drives, or keeping background processes alive. A significant chunk will choose the easier path.

Since the Foundation’s focus has shifted, I have rewriten CloudBridge to use indexd + SDK directly (no Renterd). That makes the whole thing cleaner anyway. @mecsbecs

Appreciate the back-and-forth; it sharpens the idea.

[mecsbecs]

Hi @DappMentors - thank you for this revised proposal. We’ve reached capacity for next week’s Grants Committee meeting, so this proposal will be slotted for review during the next meeting on May 26th.

Thank you for your patience.